{"schema_version":1,"updated":"2026-08-30","currency":"USD","unit":"per_1m_tokens","notes":"All prices are USD per 1,000,000 tokens unless stated otherwise. 15 of the 16 providers in this dataset were verified by fetching an official pricing page directly and reading the figures off it; the 16th, OpenRouter, was verified via its own public pricing API (see the 'official_api' verification level) -- OpenRouter is a reseller, so its prices are what you pay through OpenRouter, not necessarily the originating lab's own price. Per-model source URLs and verification dates travel with each entry. See SOURCES.md for per-provider notes and for models deliberately left out.","providers":[{"id":"anthropic","name":"Anthropic","pricing_url":"https://platform.claude.com/docs/en/about-claude/pricing","models":[{"id":"claude-opus-5","name":"Claude Opus 5","tier":"flagship","context_window":1000000,"max_output":128000,"input":5,"output":25,"cache_write":6.25,"cache_read":0.5,"modalities":["text","image"],"released":null,"verified":"2026-08-29","source":"https://platform.claude.com/docs/en/about-claude/pricing","note":"5-minute cache write shown ($6.25 = 1.25x base input); 1-hour cache write is $10/MTok (2x base). Batch API is 50% off ($2.50/$12.50). Context window and max output cross-checked against platform.claude.com/docs/en/about-claude/models/overview."},{"id":"claude-sonnet-5","name":"Claude Sonnet 5","tier":"mid","context_window":1000000,"max_output":128000,"input":2,"output":10,"cache_write":2.5,"cache_read":0.2,"modalities":["text","image"],"released":null,"verified":"2026-08-29","source":"https://platform.claude.com/docs/en/about-claude/pricing","note":"5-minute cache write shown; 1-hour cache write is $4/MTok. The page now states explicitly the previously-scheduled Sep 1 2026 increase to $3/$15 will NOT occur - $2/$10 is permanent."},{"id":"claude-haiku-4-5","name":"Claude Haiku 4.5","tier":"small","context_window":200000,"max_output":64000,"input":1,"output":5,"cache_write":1.25,"cache_read":0.1,"modalities":["text","image"],"released":null,"verified":"2026-08-29","source":"https://platform.claude.com/docs/en/about-claude/pricing","note":"5-minute cache write shown; 1-hour cache write is $2/MTok."},{"id":"claude-fable-5","name":"Claude Fable 5","tier":"reasoning","context_window":1000000,"max_output":128000,"input":10,"output":50,"cache_write":12.5,"cache_read":1,"modalities":["text","image"],"released":null,"verified":"2026-08-29","source":"https://platform.claude.com/docs/en/about-claude/pricing","note":"5-minute cache write shown; 1-hour cache write is $20/MTok."}],"verification":"official_page"},{"id":"openai","name":"OpenAI","pricing_url":"https://developers.openai.com/api/docs/pricing","models":[{"id":"gpt-5.6-sol","name":"GPT-5.6 Sol","tier":"flagship","context_window":1050000,"max_output":128000,"input":4,"output":20,"cache_write":5,"cache_read":0.4,"modalities":["text","image"],"released":"2026-07","verified":"2026-08-29","source":"https://developers.openai.com/api/docs/pricing","note":"Short-context rate, for prompts up to 272K input tokens. Above 272K the rate is $8 input / $30 output, cache read $0.80, cache write $10 per 1M. Promotional pricing, stated to run at least through 21 November 2026."},{"id":"gpt-5.6-terra","name":"GPT-5.6 Terra","tier":"mid","context_window":1050000,"max_output":128000,"input":2,"output":12,"cache_write":2.5,"cache_read":0.2,"modalities":["text","image"],"released":"2026-07","verified":"2026-08-29","source":"https://developers.openai.com/api/docs/pricing","note":"One flat rate: unlike Sol and Cyber, Terra has no long-context surcharge tier on the pricing page."},{"id":"gpt-5.6-luna","name":"GPT-5.6 Luna","tier":"small","context_window":1050000,"max_output":128000,"input":0.2,"output":1.2,"cache_write":0.25,"cache_read":0.02,"modalities":["text","image"],"released":"2026-07","verified":"2026-08-29","source":"https://developers.openai.com/api/docs/pricing","note":"One flat rate, with no long-context surcharge tier on the pricing page."},{"id":"o4-mini","name":"o4-mini","tier":"reasoning","context_window":200000,"max_output":100000,"input":1.1,"output":4.4,"cache_write":null,"cache_read":0.275,"modalities":["text","image"],"released":"2025-04","verified":"2026-08-29","source":"https://developers.openai.com/api/docs/models/o4-mini"},{"id":"gpt-5.6-cyber","name":"GPT-5.6 Cyber","tier":"reasoning","context_window":400000,"max_output":128000,"input":12.5,"output":75,"cache_write":15.625,"cache_read":1.25,"modalities":["text","image"],"released":null,"verified":"2026-08-29","source":"https://developers.openai.com/api/docs/pricing","note":"One flat rate, with no published long-context tier. Access is restricted to OpenAI's Daybreak program; also aliased as gpt-daybreak-red-latest."}],"verification":"official_page"},{"id":"google","name":"Google","pricing_url":"https://ai.google.dev/gemini-api/docs/pricing","models":[{"id":"gemini-3.1-pro","name":"Gemini 3.1 Pro","tier":"flagship","context_window":1000000,"max_output":66000,"input":2,"output":12,"cache_write":null,"cache_read":0.2,"modalities":["text","image"],"released":"2025-11","verified":"2026-08-29","source":"https://ai.google.dev/gemini-api/docs/pricing","note":"Listed as gemini-3.1-pro-preview. Standard tier for prompts up to 200K tokens; above that, $4 input / $18 output and $0.40 cache read per 1M. Cached context is also billed for storage at $4.50 per 1M tokens per hour."},{"id":"gemini-3.7-flash","name":"Gemini 3.7 Flash","tier":"mid","context_window":1000000,"max_output":64000,"input":0.75,"output":3.75,"cache_write":null,"cache_read":0.075,"modalities":["text","image"],"released":"2026-08","verified":"2026-08-29","source":"https://ai.google.dev/gemini-api/docs/pricing","note":"Standard tier. Introductory pricing through 31 December 2026; from 1 January 2027 the rate is $1.50 input / $7.50 output, cache read $0.15 per 1M. Cached context is billed for storage at $0.50 per 1M tokens per hour, rising to $1.00."},{"id":"gemini-3.6-flash","name":"Gemini 3.6 Flash","tier":"mid","context_window":1048576,"max_output":65536,"input":0.75,"output":3.75,"cache_write":null,"cache_read":0.075,"modalities":["text","image"],"released":"2026-07","verified":"2026-08-29","source":"https://ai.google.dev/gemini-api/docs/pricing","note":"Standard tier. Introductory pricing through 31 December 2026, identical to 3.7 Flash; from 1 January 2027 the rate is $1.50 input / $7.50 output, cache read $0.15 per 1M."},{"id":"gemini-3.5-flash","name":"Gemini 3.5 Flash","tier":"mid","context_window":1048576,"max_output":65536,"input":1.5,"output":9,"cache_write":null,"cache_read":0.15,"modalities":["text","image"],"released":"2026-05","verified":"2026-08-29","source":"https://ai.google.dev/gemini-api/docs/pricing","note":"Standard tier, not introductory pricing. Cached context is billed for storage at $1 per 1M tokens per hour."},{"id":"gemini-3.1-flash-lite","name":"Gemini 3.1 Flash-Lite","tier":"small","context_window":1048576,"max_output":null,"input":0.25,"output":1.5,"cache_write":null,"cache_read":0.025,"modalities":["text","image"],"released":"2026-05","verified":"2026-08-29","source":"https://ai.google.dev/gemini-api/docs/pricing","note":"Rate shown covers text, image and video input. Audio input is $0.50 per 1M, audio caching $0.05 per 1M. Cached context is billed for storage at $1 per 1M tokens per hour."},{"id":"gemini-3.5-flash-lite","name":"Gemini 3.5 Flash-Lite","tier":"small","context_window":1048576,"max_output":65536,"input":0.3,"output":2.5,"cache_write":null,"cache_read":0.03,"modalities":["text","image"],"released":null,"source":"https://ai.google.dev/gemini-api/docs/pricing","note":"Standard tier. One input rate covers text, image, video and audio. Cached context is billed for storage at $1 per 1M tokens per hour.","verified":"2026-08-29"},{"id":"gemini-3-flash-preview","name":"Gemini 3 Flash Preview","tier":"small","context_window":1048576,"max_output":65536,"input":0.5,"output":3,"cache_write":null,"cache_read":0.05,"modalities":["text","image"],"released":null,"source":"https://ai.google.dev/gemini-api/docs/pricing","note":"Standard tier, text/image/video rate; audio input $1.00/1M. Cache storage $1.00/1M tokens/hour. Earlier Gemini 3 Flash preview release, not marked deprecated on the page.","verified":"2026-08-29"},{"id":"gemini-2.5-pro","name":"Gemini 2.5 Pro","tier":"flagship","context_window":1048576,"max_output":65536,"input":1.25,"output":10,"cache_write":null,"cache_read":0.125,"modalities":["text","image"],"released":null,"source":"https://ai.google.dev/gemini-api/docs/pricing","note":"Standard tier, <=200k tokens. >200k tier: input $2.50, output $15.00, cache $0.25/1M; cache storage $4.50/1M tokens/hour. Previous-generation flagship, still fully listed (not deprecated) on the current pricing page -- important reference point for price comparisons.","verified":"2026-08-29"},{"id":"gemini-2.5-flash","name":"Gemini 2.5 Flash","tier":"mid","context_window":1000000,"max_output":null,"input":0.3,"output":2.5,"cache_write":null,"cache_read":0.03,"modalities":["text","image"],"released":null,"source":"https://ai.google.dev/gemini-api/docs/pricing","note":"Standard tier, text/image/video rate; audio input $1.00/1M. Cache storage $1.00/1M tokens/hour. Page explicitly states this model supports a 1M token context window.","verified":"2026-08-29"},{"id":"gemini-2.5-flash-lite","name":"Gemini 2.5 Flash-Lite","tier":"small","context_window":1048576,"max_output":65536,"input":0.1,"output":0.4,"cache_write":null,"cache_read":0.01,"modalities":["text","image"],"released":null,"source":"https://ai.google.dev/gemini-api/docs/pricing","note":"Standard tier, text/image/video rate; audio input $0.30/1M. Cache storage $1.00/1M tokens/hour.","verified":"2026-08-29"}],"verification":"official_page"},{"id":"mistral","name":"Mistral AI","pricing_url":"https://mistral.ai/pricing/api","models":[{"id":"mistral-large-3","name":"Mistral Large 3","tier":"flagship","context_window":256000,"max_output":null,"input":0.5,"output":1.5,"cache_write":null,"cache_read":null,"modalities":["text"],"released":"2025-12","verified":"2026-08-29","source":"https://mistral.ai/pricing"},{"id":"mistral-medium-3.5","name":"Mistral Medium 3.5","tier":"mid","context_window":256000,"max_output":null,"input":1.5,"output":7.5,"cache_write":null,"cache_read":null,"modalities":["text"],"released":"2026-04","verified":"2026-08-29","source":"https://mistral.ai/pricing"},{"id":"mistral-small-4","name":"Mistral Small 4","tier":"small","context_window":256000,"max_output":null,"input":0.15,"output":0.6,"cache_write":null,"cache_read":null,"modalities":["text"],"released":"2026-03","verified":"2026-08-29","source":"https://mistral.ai/news/mistral-small-4/"},{"id":"codestral","name":"Codestral","tier":"small","context_window":128000,"max_output":null,"input":0.3,"output":0.9,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-29","source":"https://docs.mistral.ai/models/codestral"},{"id":"ministral-3-14b","name":"Ministral 3 14B","tier":"small","context_window":256000,"max_output":null,"input":0.2,"output":0.2,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-29","source":"https://docs.mistral.ai/models/ministral-3-14b-25-12"},{"id":"ministral-3-8b","name":"Ministral 3 8B","tier":"small","context_window":256000,"max_output":null,"input":0.15,"output":0.15,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-29","source":"https://docs.mistral.ai/models/ministral-3-8b-25-12"},{"id":"ministral-3-3b","name":"Ministral 3 3B","tier":"small","context_window":256000,"max_output":null,"input":0.1,"output":0.1,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-29","source":"https://docs.mistral.ai/models/ministral-3-3b-25-12"},{"id":"glm-5.2","name":"GLM-5.2 (on Mistral)","tier":"flagship","context_window":1000000,"max_output":128000,"input":1.4,"output":4.4,"cache_write":null,"cache_read":0.14,"modalities":["text"],"released":null,"verified":"2026-08-29","source":"https://docs.mistral.ai/models/zai-glm-5-2","note":"A Z.ai model resold by Mistral and served on Mistral's EU infrastructure, listed as zai-glm-5-2 on Mistral's own pricing page. Cache read is $0.14 per 1M."}],"verification":"official_page"},{"id":"deepseek","name":"DeepSeek","pricing_url":"https://api-docs.deepseek.com/quick_start/pricing","models":[{"id":"deepseek-v4-pro","name":"DeepSeek V4 Pro","tier":"flagship","context_window":1000000,"max_output":384000,"input":1.32,"output":3.96,"cache_write":null,"cache_read":0.044,"modalities":["text"],"released":"2026-04","verified":"2026-08-30","source":"https://api-docs.deepseek.com/quick_start/pricing","note":"Peak cache-miss rate, which is this dataset's standard tier. DeepSeek's own page states off-peak rates are half the peak rates, with peak hours 01:00-04:00 and 06:00-10:00 UTC Monday to Friday: off-peak is $0.66 input / $1.98 output, cache-hit input $0.022. The off-peak half-price is a time-of-day discount and is excluded here for the same reason batch and cached-input discounts are.","family":"deepseek-v4-pro"},{"id":"deepseek-v4-flash","name":"DeepSeek V4 Flash","tier":"small","context_window":1000000,"max_output":384000,"input":0.44,"output":1.32,"cache_write":null,"cache_read":0.014,"modalities":["text"],"released":"2026-04","verified":"2026-08-30","source":"https://api-docs.deepseek.com/quick_start/pricing","note":"Peak cache-miss rate, this dataset's standard tier. Off-peak is half: $0.22 input / $0.66 output, cache-hit input $0.007. Peak hours are 01:00-04:00 and 06:00-10:00 UTC Monday to Friday per DeepSeek's own page."},{"id":"deepseek-v4-flash-vision-exp","name":"DeepSeek V4 Flash Vision (Exp)","tier":"small","context_window":1000000,"max_output":384000,"input":0.44,"output":1.32,"cache_write":null,"cache_read":0.014,"modalities":["text","vision"],"released":null,"source":"https://api-docs.deepseek.com/quick_start/pricing","note":"Peak cache-miss rate, this dataset's standard tier; off-peak is half ($0.22/$0.66). Experimental vision-input variant of V4 Flash with the same token pricing (images billed as input tokens per DeepSeek's vision token-conversion rule). Not image-only -- a text-generation model with added image input, so in scope.","verified":"2026-08-30"}],"verification":"official_page"},{"id":"xai","name":"xAI","pricing_url":"https://docs.x.ai/developers/pricing","models":[{"id":"grok-4.6","name":"Grok 4.6","tier":"flagship","context_window":500000,"max_output":null,"input":2,"output":6,"cache_write":null,"cache_read":0.5,"modalities":["text","image"],"released":"2026-06","verified":"2026-08-29","source":"https://docs.x.ai/docs/models","note":"Short-context rate, for prompts under 200K tokens. At 200K and above: $4 input / $12 output, cache read $1 per 1M. xAI's Priority Processing tier doubles every rate and applies to all models — it is not a separate model."},{"id":"grok-4.3","name":"Grok 4.3","tier":"mid","context_window":1000000,"max_output":null,"input":1.25,"output":2.5,"cache_write":null,"cache_read":0.2,"modalities":["text","image"],"released":null,"verified":"2026-08-29","source":"https://docs.x.ai/docs/models","note":"Short-context rate, for prompts under 200K tokens. At 200K and above: $2.50 input / $5 output, cache read $0.40 per 1M. Eligible for a 20% Batch API discount."},{"id":"grok-4.5","name":"Grok 4.5","tier":"flagship","context_window":500000,"max_output":null,"input":2,"output":6,"cache_write":null,"cache_read":0.3,"modalities":["text","image"],"released":null,"source":"https://docs.x.ai/developers/pricing","note":"Short-context rate, for prompts under 200K tokens. At 200K and above: $4 input / $12 output, cache read $0.60 per 1M. Same input and output price as Grok 4.6 with a cheaper cache read; the previous flagship, not marked deprecated.","verified":"2026-08-29"},{"id":"grok-build-0.1","name":"Grok Build","tier":"small","context_window":256000,"max_output":null,"input":1,"output":2,"cache_write":null,"cache_read":0.2,"modalities":["text"],"released":null,"source":"https://docs.x.ai/developers/pricing","note":"Backs the Grok Build coding CLI.","verified":"2026-08-29"}],"verification":"official_page"},{"id":"amazon","name":"Amazon (Nova)","pricing_url":"https://aws.amazon.com/bedrock/pricing/","models":[{"id":"amazon-nova-premier","name":"Amazon Nova Premier","tier":"flagship","context_window":1000000,"max_output":10000,"input":2.5,"output":12.5,"cache_write":null,"cache_read":0.625,"modalities":["text","image"],"released":"2025-10","verified":"2026-08-29","source":"https://aws.amazon.com/bedrock/pricing/","note":"On-demand rate for US East (N. Virginia)."},{"id":"amazon-nova-pro","name":"Amazon Nova Pro","tier":"mid","context_window":300000,"max_output":10000,"input":0.8,"output":3.2,"cache_write":0,"cache_read":0.2,"modalities":["text","image"],"released":"2024-12","verified":"2026-08-29","source":"https://aws.amazon.com/bedrock/pricing/","note":"On-demand rate for US East (N. Virginia). Cache writes are billed at $0 per 1M."},{"id":"amazon-nova-lite","name":"Amazon Nova Lite","tier":"small","context_window":300000,"max_output":10000,"input":0.06,"output":0.24,"cache_write":0,"cache_read":0.015,"modalities":["text","image"],"released":"2024-12","verified":"2026-08-29","source":"https://aws.amazon.com/bedrock/pricing/","note":"On-demand rate for US East (N. Virginia). Cache writes are billed at $0 per 1M."},{"id":"amazon-nova-micro","name":"Amazon Nova Micro","tier":"small","context_window":128000,"max_output":10000,"input":0.035,"output":0.14,"cache_write":0,"cache_read":0.00875,"modalities":["text"],"released":"2024-12","verified":"2026-08-29","source":"https://aws.amazon.com/bedrock/pricing/","note":"On-demand rate for US East (N. Virginia). Cache writes are billed at $0 per 1M."},{"id":"amazon-nova-2-lite","name":"Amazon Nova 2 Lite","tier":"small","context_window":1000000,"max_output":65536,"input":0.3,"output":2.5,"cache_write":0,"cache_read":0.075,"modalities":["text","image","video"],"released":"2025-12","verified":"2026-08-29","source":"https://aws.amazon.com/bedrock/pricing/","note":"Cross-region global on-demand rate for US East (N. Virginia). Standard service tier; Priority and Flex are priced differently. A hybrid reasoning model — reasoning tokens are billed as output tokens."}],"verification":"official_page"},{"id":"cohere","name":"Cohere","pricing_url":"https://cohere.com/pricing","models":[{"id":"command-r","name":"Command R","tier":"mid","context_window":128000,"max_output":4000,"input":0.15,"output":0.6,"cache_write":null,"cache_read":null,"modalities":["text"],"released":"2024-03","verified":"2026-08-29","source":"https://docs.cohere.com/docs/command-r"},{"id":"command-r7b","name":"Command R7B","tier":"small","context_window":128000,"max_output":4000,"input":0.0375,"output":0.15,"cache_write":null,"cache_read":null,"modalities":["text"],"released":"2024-12","verified":"2026-08-29","source":"https://docs.cohere.com/docs/command-r7b"}],"verification":"official_page"},{"id":"groq","name":"Groq","pricing_url":"https://console.groq.com/docs/models","verification":"official_page","models":[{"id":"groq-gpt-oss-120b","name":"GPT-OSS 120B","tier":"flagship","context_window":131072,"max_output":65536,"input":0.15,"output":0.6,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-29","source":"https://console.groq.com/docs/models","note":"OpenAI open-weight reasoning model on Groq. Table header explicitly states 'PRICE PER 1M TOKENS'.","family":"gpt-oss-120b"},{"id":"groq-gpt-oss-20b","name":"GPT-OSS 20B","tier":"small","context_window":131072,"max_output":65536,"input":0.075,"output":0.3,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-29","source":"https://console.groq.com/docs/models","note":"Cheapest general-purpose text model with public pricing on Groq's models page.","family":"gpt-oss-20b"},{"id":"groq-qwen3-6-27b","name":"Qwen3.6 27B","tier":"mid","context_window":131072,"max_output":16384,"input":0.6,"output":3,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-29","source":"https://console.groq.com/docs/models","note":"Listed as qwen/qwen3.6-27b in Groq's model catalog table."},{"id":"groq-qwen3-8-27b","name":"Qwen3.8 27B","tier":"mid","context_window":131042,"max_output":16384,"input":0.8,"output":4,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-29","source":"https://console.groq.com/docs/models","note":"Listed as qwen/qwen3.8-27b. Groq publishes the context window as 131,042 tokens, slightly below the 131,072 it shows for its other models."}]},{"id":"fireworks","name":"Fireworks AI","pricing_url":"https://fireworks.ai/pricing","verification":"official_page","models":[{"id":"fireworks-kimi-k3","name":"Kimi K3","tier":"flagship","context_window":null,"max_output":null,"input":3,"output":15,"cache_write":null,"cache_read":0.3,"modalities":["text"],"released":null,"verified":"2026-08-29","source":"https://docs.fireworks.ai/serverless/pricing","note":"Standard-tier serverless price (input / cached input / output, USD per 1M tokens, as stated on the page). A pricier Priority tier and 'Kimi K3 Fast' variant also exist on the same page.","family":"kimi-k3"},{"id":"fireworks-deepseek-v4-pro","name":"DeepSeek V4 Pro","tier":"flagship","context_window":null,"max_output":null,"input":1.74,"output":3.48,"cache_write":null,"cache_read":0.145,"modalities":["text"],"released":null,"verified":"2026-08-29","source":"https://docs.fireworks.ai/serverless/pricing","note":"Standard-tier serverless price. Page also lists a 'DeepSeek V4 Pro (0813)' dated snapshot at a different price.","family":"deepseek-v4-pro"},{"id":"fireworks-qwen3-8-max","name":"Qwen3.8 Max","tier":"mid","context_window":null,"max_output":null,"input":2,"output":6,"cache_write":null,"cache_read":0.25,"modalities":["text"],"released":null,"verified":"2026-08-29","source":"https://docs.fireworks.ai/serverless/pricing","note":"Standard-tier serverless price.","family":"qwen3.8-max"},{"id":"fireworks-glm-5-3","name":"GLM 5.3","tier":"mid","context_window":null,"max_output":null,"input":1.4,"output":4.4,"cache_write":null,"cache_read":0.26,"modalities":["text"],"released":null,"verified":"2026-08-29","source":"https://docs.fireworks.ai/serverless/pricing","note":"Standard-tier serverless price. Distinctive model family prominently hosted by Fireworks.","family":"glm-5.3"},{"id":"fireworks-gpt-oss-120b","name":"GPT-OSS 120B","tier":"mid","context_window":131072,"max_output":null,"input":0.15,"output":0.6,"cache_write":null,"cache_read":0.015,"modalities":["text"],"released":null,"verified":"2026-08-29","source":"https://docs.fireworks.ai/serverless/pricing","note":"Standard-tier serverless price.","family":"gpt-oss-120b"},{"id":"fireworks-gpt-oss-20b","name":"GPT-OSS 20B","tier":"small","context_window":131072,"max_output":null,"input":0.07,"output":0.3,"cache_write":null,"cache_read":0.035,"modalities":["text"],"released":null,"verified":"2026-08-29","source":"https://docs.fireworks.ai/serverless/pricing","note":"Cheapest usable small model listed. No Priority tier offered for this model per the page.","family":"gpt-oss-20b"}]},{"id":"deepinfra","name":"DeepInfra","pricing_url":"https://deepinfra.com/pricing","verification":"official_page","models":[{"id":"deepinfra-deepseek-v4-pro","name":"DeepSeek-V4-Pro","tier":"flagship","context_window":1024000,"max_output":null,"input":1.3,"output":2.6,"cache_write":null,"cache_read":0.1,"modalities":["text"],"released":null,"verified":"2026-08-29","source":"https://deepinfra.com/pricing","note":"Context listed on page as '1024k'.","family":"deepseek-v4-pro"},{"id":"deepinfra-kimi-k3","name":"Kimi-K3","tier":"flagship","context_window":1024000,"max_output":null,"input":2.85,"output":14.25,"cache_write":null,"cache_read":0.285,"modalities":["text"],"released":null,"verified":"2026-08-29","source":"https://deepinfra.com/pricing","note":"Context listed on page as '1024k'.","family":"kimi-k3"},{"id":"deepinfra-llama-4-maverick-17b-128e","name":"Llama-4-Maverick-17B-128E","tier":"mid","context_window":1024000,"max_output":null,"input":0.2,"output":0.8,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-29","source":"https://deepinfra.com/pricing","note":"Context listed on page as '1024k'. No cached-input rate shown for this model.","family":"llama-4-maverick"},{"id":"deepinfra-qwen3-max","name":"Qwen3-Max","tier":"mid","context_window":250000,"max_output":null,"input":1.2,"output":6,"cache_write":null,"cache_read":0.24,"modalities":["text"],"released":null,"verified":"2026-08-29","source":"https://deepinfra.com/pricing","note":"Context listed on page as '250k'."},{"id":"deepinfra-llama-3.3-70b-instruct-turbo","name":"Llama-3.3-70B-Instruct-Turbo","tier":"mid","context_window":128000,"max_output":null,"input":0.1,"output":0.32,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-29","source":"https://deepinfra.com/pricing","note":"Widely used open-weights model; context listed on page as '128k'.","family":"llama-3.3-70b"},{"id":"deepinfra-llama-3.1-8b-instruct-turbo","name":"Meta-Llama-3.1-8B-Instruct-Turbo","tier":"small","context_window":128000,"max_output":null,"input":0.02,"output":0.04,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-29","source":"https://deepinfra.com/pricing","note":"Cheapest usable small model listed; context on page shown as '128k'.","family":"llama-3.1-8b"}]},{"id":"alibaba","name":"Alibaba Qwen","pricing_url":"https://www.alibabacloud.com/help/en/model-studio/model-pricing","verification":"official_page","models":[{"id":"qwen3.8-max","name":"Qwen3.8-Max","tier":"flagship","context_window":1000000,"max_output":null,"input":2,"output":6,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-29","source":"https://www.alibabacloud.com/help/en/model-studio/model-pricing","note":"International (Singapore) USD pricing, flat rate 0<tokens<=1M.","family":"qwen3.8-max"},{"id":"qwen-plus","name":"Qwen-Plus","tier":"mid","context_window":1000000,"max_output":null,"input":0.4,"output":1.2,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-29","source":"https://www.alibabacloud.com/help/en/model-studio/model-pricing","note":"Tiered by context length: 0-256K shown ($0.4/$1.2 per 1M); 256K-1M tier is $1.2/$3.6 per 1M."},{"id":"qwen3.5-flash","name":"Qwen3.5-Flash","tier":"small","context_window":1000000,"max_output":null,"input":0.1,"output":0.4,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-29","source":"https://www.alibabacloud.com/help/en/model-studio/model-pricing","note":"International USD pricing, flat rate 0<tokens<=1M."},{"id":"qwen-turbo","name":"Qwen-Turbo","tier":"small","context_window":1000000,"max_output":null,"input":0.05,"output":0.2,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-29","source":"https://www.alibabacloud.com/help/en/model-studio/model-pricing","note":"Cheapest Qwen chat model listed; no tiered pricing shown on page."},{"id":"qwen3.8-flash","name":"Qwen3.8-Flash","tier":"small","context_window":1000000,"max_output":null,"input":0.15,"output":0.47,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-29","source":"https://www.alibabacloud.com/help/en/model-studio/model-pricing","note":"International USD pricing, flat rate 0<tokens<=1M."}]},{"id":"moonshot","name":"Moonshot AI","pricing_url":"https://platform.kimi.ai/docs/pricing","verification":"official_page","models":[{"id":"kimi-k3","name":"Kimi K3","tier":"flagship","context_window":1048576,"max_output":null,"input":3,"output":15,"cache_write":null,"cache_read":0.3,"modalities":["text"],"released":null,"verified":"2026-08-29","source":"https://platform.kimi.ai/docs/pricing/chat-k3","note":"input is cache-miss (standard) price per 1M tokens; cache_read is cache-hit price ($0.30).","family":"kimi-k3"},{"id":"kimi-k2.6","name":"Kimi K2.6","tier":"mid","context_window":262144,"max_output":null,"input":0.95,"output":4,"cache_write":null,"cache_read":0.16,"modalities":["text"],"released":null,"verified":"2026-08-29","source":"https://platform.kimi.ai/docs/pricing/chat-k26","note":"input is cache-miss (standard) price per 1M tokens; cache_read is cache-hit price ($0.16)."},{"id":"kimi-k2.7-code","name":"Kimi K2.7 Code","tier":"mid","context_window":262144,"max_output":null,"input":0.95,"output":4,"cache_write":null,"cache_read":0.19,"modalities":["text"],"released":null,"verified":"2026-08-29","source":"https://platform.kimi.ai/docs/pricing/chat-k27-code","note":"Coding-focused model; input is cache-miss price, cache_read is cache-hit price ($0.19). A HighSpeed variant costs $1.90/$8.00."},{"id":"moonshot-v1-128k","name":"Moonshot v1 128K","tier":"small","context_window":131072,"max_output":null,"input":2,"output":5,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-29","source":"https://platform.kimi.ai/docs/pricing/chat-v1","note":"Legacy generation model, per 1M tokens."},{"id":"moonshot-v1-8k","name":"Moonshot v1 8K","tier":"small","context_window":8192,"max_output":null,"input":0.2,"output":2,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-29","source":"https://platform.kimi.ai/docs/pricing/chat-v1","note":"Cheapest legacy Moonshot v1 tier, per 1M tokens."}]},{"id":"zai","name":"Z.ai","pricing_url":"https://docs.z.ai/guides/overview/pricing","verification":"official_page","models":[{"id":"glm-5.3","name":"GLM-5.3","tier":"flagship","context_window":1000000,"max_output":128000,"input":1.4,"output":4.4,"cache_write":null,"cache_read":0.26,"modalities":["text"],"released":null,"verified":"2026-08-29","source":"https://docs.z.ai/guides/overview/pricing","note":"Latest flagship text model, per 1M tokens. Context window not stated on pricing page.","family":"glm-5.3"},{"id":"glm-5.3-flash","name":"GLM-5.3-Flash","tier":"small","context_window":1000000,"max_output":128000,"input":0.075,"output":0.25,"cache_write":null,"cache_read":0.015,"modalities":["text"],"released":null,"verified":"2026-08-29","source":"https://docs.z.ai/guides/overview/pricing","note":"Promotional price (50% off list of $0.15/$0.50) shown on page through 2026-09-09; list price may apply after that date."},{"id":"glm-4.6","name":"GLM-4.6","tier":"mid","context_window":200000,"max_output":128000,"input":0.6,"output":2.2,"cache_write":null,"cache_read":0.11,"modalities":["text"],"released":null,"verified":"2026-08-29","source":"https://docs.z.ai/guides/overview/pricing","note":"Widely-used popular GLM generation, per 1M tokens."},{"id":"glm-4.5","name":"GLM-4.5","tier":"mid","context_window":128000,"max_output":96000,"input":0.6,"output":2.2,"cache_write":null,"cache_read":0.11,"modalities":["text"],"released":null,"verified":"2026-08-29","source":"https://docs.z.ai/guides/overview/pricing","note":"Popular open-weight GLM generation, per 1M tokens."},{"id":"glm-4.5-air","name":"GLM-4.5-Air","tier":"small","context_window":128000,"max_output":null,"input":0.2,"output":1.1,"cache_write":null,"cache_read":0.03,"modalities":["text"],"released":null,"verified":"2026-08-29","source":"https://docs.z.ai/guides/overview/pricing","note":"Cheapest non-free usable GLM text model, per 1M tokens."}]},{"id":"openrouter","name":"OpenRouter","pricing_url":"https://openrouter.ai/models","models":[{"id":"openrouter-deepseek-v4-pro","name":"DeepSeek V4 Pro","tier":"flagship","context_window":1048576,"max_output":384000,"input":0.680862,"output":1.361724,"cache_write":null,"cache_read":0.056739,"modalities":["text"],"released":null,"verified":"2026-08-29","source":"https://openrouter.ai/deepseek/deepseek-v4-pro","note":"OpenRouter's default route for this model. DeepSeek itself is not one of the backends OpenRouter lists for it, so this is a third-party host's price, not DeepSeek's own -- compare against DeepSeek's own direct pricing if that distinction matters for you.","family":"deepseek-v4-pro"},{"id":"openrouter-glm-5.3","name":"GLM-5.3","tier":"flagship","context_window":1310720,"max_output":131072,"input":1.4,"output":4.4,"cache_write":null,"cache_read":0.26,"modalities":["text"],"released":null,"verified":"2026-08-29","source":"https://openrouter.ai/z-ai/glm-5.3","note":"OpenRouter's default route for this model, currently Z.ai's own listed price.","family":"glm-5.3"},{"id":"openrouter-gpt-oss-120b","name":"GPT-OSS 120B","tier":"mid","context_window":131072,"max_output":117964,"input":0.037,"output":0.17,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-29","source":"https://openrouter.ai/openai/gpt-oss-120b","note":"OpenRouter's default route for this model; several dozen independent backends host it at different prices, this is only the one OpenRouter routes to by default.","family":"gpt-oss-120b"},{"id":"openrouter-gpt-oss-20b","name":"GPT-OSS 20B","tier":"small","context_window":131072,"max_output":117964,"input":0.03,"output":0.13,"cache_write":null,"cache_read":0.03,"modalities":["text"],"released":null,"verified":"2026-08-29","source":"https://openrouter.ai/openai/gpt-oss-20b","note":"OpenRouter's default route for this model.","family":"gpt-oss-20b"},{"id":"openrouter-kimi-k3","name":"Kimi K3","tier":"flagship","context_window":1048576,"max_output":943718,"input":3,"output":15,"cache_write":null,"cache_read":0.3,"modalities":["text","image","video"],"released":null,"verified":"2026-08-29","source":"https://openrouter.ai/moonshotai/kimi-k3","note":"OpenRouter's default route for this model, currently Moonshot AI's own listed price.","family":"kimi-k3"},{"id":"openrouter-llama-3.1-8b","name":"Llama 3.1 8B","tier":"small","context_window":131072,"max_output":117964,"input":0.05,"output":0.08,"cache_write":null,"cache_read":0.025,"modalities":["text"],"released":null,"verified":"2026-08-29","source":"https://openrouter.ai/meta-llama/llama-3.1-8b-instruct","note":"OpenRouter's default route for this model.","family":"llama-3.1-8b"},{"id":"openrouter-llama-3.3-70b","name":"Llama 3.3 70B","tier":"mid","context_window":131072,"max_output":115200,"input":0.71,"output":0.71,"cache_write":null,"cache_read":0.71,"modalities":["text"],"released":null,"verified":"2026-08-29","source":"https://openrouter.ai/meta-llama/llama-3.3-70b-instruct","note":"OpenRouter's default route for this model.","family":"llama-3.3-70b"},{"id":"openrouter-llama-4-maverick","name":"Llama 4 Maverick","tier":"mid","context_window":1048576,"max_output":16384,"input":0.2,"output":0.8,"cache_write":null,"cache_read":null,"modalities":["text","image"],"released":null,"verified":"2026-08-29","source":"https://openrouter.ai/meta-llama/llama-4-maverick","note":"OpenRouter's default route for this model, currently DeepInfra's price.","family":"llama-4-maverick"},{"id":"openrouter-qwen3.8-max","name":"Qwen3.8-Max","tier":"flagship","context_window":1000000,"max_output":131072,"input":2,"output":6,"cache_write":2.5,"cache_read":0.25,"modalities":["text","image","video"],"released":null,"verified":"2026-08-29","source":"https://openrouter.ai/qwen/qwen3.8-max","note":"OpenRouter's default route for this model, currently Alibaba's own listed price.","family":"qwen3.8-max"},{"id":"openrouter-nemotron-3-super","name":"Nemotron 3 Super","tier":"mid","context_window":1000000,"max_output":16384,"input":0.085,"output":0.4,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-29","source":"https://openrouter.ai/nvidia/nemotron-3-super-120b-a12b","note":"Nvidia's open-weight reasoning-capable MoE model, mixture-of-experts with 12B active parameters."},{"id":"openrouter-sonar","name":"Sonar","tier":"small","context_window":127072,"max_output":114364,"input":1,"output":1,"cache_write":null,"cache_read":null,"modalities":["text","image"],"released":null,"verified":"2026-08-29","source":"https://openrouter.ai/perplexity/sonar","note":"Perplexity's search-augmented model; answers are grounded in live web results, which is the reason to reach for it over a plain chat model, not just the price."},{"id":"openrouter-sonar-pro","name":"Sonar Pro","tier":"flagship","context_window":200000,"max_output":8000,"input":3,"output":15,"cache_write":null,"cache_read":null,"modalities":["text","image"],"released":null,"verified":"2026-08-29","source":"https://openrouter.ai/perplexity/sonar-pro","note":"Perplexity's flagship search-augmented model. Note the small max output relative to context window."},{"id":"openrouter-ernie-4.5-vl","name":"ERNIE 4.5 VL 424B A47B","tier":"mid","context_window":123000,"max_output":16000,"input":0.42,"output":1.25,"cache_write":null,"cache_read":null,"modalities":["text","image"],"released":null,"verified":"2026-08-29","source":"https://openrouter.ai/baidu/ernie-4.5-vl-424b-a47b","note":"Baidu's vision-language MoE model, 424B total / 47B active parameters."},{"id":"openrouter-hy3","name":"Hy3","tier":"mid","context_window":262144,"max_output":128000,"input":0.132,"output":0.528,"cache_write":null,"cache_read":0.033,"modalities":["text"],"released":null,"verified":"2026-08-29","source":"https://openrouter.ai/tencent/hy3","note":"Tencent's Hunyuan model line, text-only."},{"id":"openrouter-minimax-m3","name":"MiniMax M3","tier":"flagship","context_window":1048576,"max_output":512000,"input":0.3,"output":1.2,"cache_write":null,"cache_read":0.06,"modalities":["text","image","video"],"released":null,"verified":"2026-08-29","source":"https://openrouter.ai/minimax/minimax-m3","note":"MiniMax's current flagship model, with a 1M-token context window."},{"id":"openrouter-seed-2.0-lite","name":"Seed 2.0 Lite","tier":"small","context_window":262144,"max_output":131072,"input":0.25,"output":2,"cache_write":null,"cache_read":null,"modalities":["text","image","video"],"released":null,"verified":"2026-08-29","source":"https://openrouter.ai/bytedance-seed/seed-2.0-lite","note":"ByteDance's Seed (Doubao) model line, supports text, image and video input."},{"id":"openrouter-mimo-v2.5","name":"MiMo V2.5","tier":"small","context_window":1050000,"max_output":131072,"input":0.14,"output":0.28,"cache_write":null,"cache_read":0.0028,"modalities":["text","audio","image","video"],"released":null,"verified":"2026-08-29","source":"https://openrouter.ai/xiaomi/mimo-v2.5","note":"Xiaomi's MiMo line, with text, audio, image and video input support."},{"id":"openrouter-granite-4.1-8b","name":"Granite 4.1 8B","tier":"small","context_window":131072,"max_output":117964,"input":0.05,"output":0.1,"cache_write":null,"cache_read":0.05,"modalities":["text"],"released":null,"verified":"2026-08-29","source":"https://openrouter.ai/ibm-granite/granite-4.1-8b","note":"IBM's enterprise-oriented open-weight model line."},{"id":"openrouter-solar-pro-4","name":"Solar Pro 4","tier":"small","context_window":524288,"max_output":131072,"input":0.03,"output":0.12,"cache_write":null,"cache_read":0.006,"modalities":["text"],"released":null,"verified":"2026-08-29","source":"https://openrouter.ai/upstage/solar-pro4","note":"Upstage's Solar model line."}],"verification":"official_api"},{"id":"together","name":"Together AI","pricing_url":"https://www.together.ai/pricing","verification":"official_page","models":[{"id":"together-kimi-k3","name":"Kimi K3","tier":"flagship","context_window":null,"max_output":null,"input":3,"output":15,"cache_write":null,"cache_read":0.3,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://www.together.ai/pricing","note":"Serverless Chat endpoint on Together AI. Cached input tokens billed at $0.30/M."},{"id":"together-deepseek-v4-flash-0731","name":"DeepSeek V4 Flash 0731","tier":"small","context_window":null,"max_output":null,"input":0.14,"output":0.28,"cache_write":null,"cache_read":0.03,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://www.together.ai/pricing","note":"Serverless Chat endpoint on Together AI. Cached input tokens billed at $0.03/M."},{"id":"together-deepseek-v4-pro-0813","name":"DeepSeek V4 Pro 0813","tier":"flagship","context_window":null,"max_output":null,"input":1.32,"output":3.96,"cache_write":null,"cache_read":0.13,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://www.together.ai/pricing","note":"Serverless Chat endpoint on Together AI. Cached input tokens billed at $0.13/M."},{"id":"together-glm-5-3","name":"GLM-5.3","tier":"flagship","context_window":null,"max_output":null,"input":1.4,"output":4.4,"cache_write":null,"cache_read":0.26,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://www.together.ai/pricing","note":"Serverless Chat endpoint on Together AI. Cached input tokens billed at $0.26/M."},{"id":"together-glm-5-3-flash","name":"GLM-5.3-Flash","tier":"small","context_window":null,"max_output":null,"input":0.15,"output":0.5,"cache_write":null,"cache_read":0.03,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://www.together.ai/pricing","note":"Serverless Chat endpoint on Together AI. Cached input tokens billed at $0.03/M."},{"id":"together-gpt-oss-120b","name":"gpt-oss-120B","tier":"mid","context_window":null,"max_output":null,"input":0.15,"output":0.6,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://www.together.ai/pricing","note":"Serverless Chat endpoint on Together AI."},{"id":"together-gpt-oss-20b","name":"gpt-oss-20B","tier":"small","context_window":null,"max_output":null,"input":0.05,"output":0.2,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://www.together.ai/pricing","note":"Serverless Chat endpoint on Together AI."},{"id":"together-qwen3-5-397b-a17b","name":"Qwen3.5-397B-A17B","tier":"flagship","context_window":null,"max_output":null,"input":0.6,"output":3.6,"cache_write":null,"cache_read":0.35,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://www.together.ai/pricing","note":"Serverless Chat endpoint on Together AI. Cached input tokens billed at $0.35/M."},{"id":"together-qwen3-235b-a22b-instruct-2507","name":"Qwen3 235B A22B Instruct 2507 FP8 Throughput","tier":"mid","context_window":null,"max_output":null,"input":0.2,"output":0.6,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://www.together.ai/pricing","note":"Serverless FP8 Throughput endpoint on Together AI."},{"id":"together-minimax-m3","name":"MiniMax M3","tier":"mid","context_window":null,"max_output":null,"input":0.3,"output":1.2,"cache_write":null,"cache_read":0.06,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://www.together.ai/pricing","note":"Serverless Chat endpoint on Together AI. Cached input tokens billed at $0.06/M."},{"id":"together-llama-4-maverick","name":"Llama 4 Maverick","tier":"flagship","context_window":1048000,"max_output":null,"input":0.27,"output":0.85,"cache_write":null,"cache_read":null,"modalities":["text","image"],"released":"2025-04","verified":"2026-08-29","source":"https://www.together.ai/models/llama-4-maverick","note":"Together's model page gives two different context lengths: 1,048,576 tokens in its spec table and 524,288 in its description. The spec table is used here.","family":"llama-4-maverick"},{"id":"together-llama-4-scout","name":"Llama 4 Scout","tier":"mid","context_window":1000000,"max_output":null,"input":0.18,"output":0.59,"cache_write":null,"cache_read":null,"modalities":["text","image"],"released":"2025-04","verified":"2026-08-29","source":"https://www.together.ai/models/llama-4-scout","note":"Priced on its own model page rather than in the main serverless pricing table."},{"id":"together-llama-3.1-8b-instruct-turbo","name":"Llama 3.1 8B","tier":"small","context_window":128000,"max_output":null,"input":0.18,"output":0.18,"cache_write":null,"cache_read":null,"modalities":["text"],"released":"2024-07","verified":"2026-08-29","source":"https://www.together.ai/pricing","note":"Flat rate — input and output cost the same. Priced on its own model page rather than in the main serverless table.","family":"llama-3.1-8b"},{"id":"together-llama-3.3-70b","name":"Llama 3.3 70B","tier":"mid","context_window":128000,"max_output":null,"input":1.04,"output":1.04,"cache_write":null,"cache_read":null,"modalities":["text"],"released":"2024-12","verified":"2026-08-30","source":"https://www.together.ai/pricing","note":"Flat rate — input and output cost the same.","family":"llama-3.3-70b"},{"id":"together-llama-3-8b-instruct-lite","name":"Llama 3 8B Instruct Lite","tier":"small","context_window":8000,"max_output":null,"input":0.14,"output":0.14,"cache_write":null,"cache_read":null,"modalities":["text"],"released":"2024-04","verified":"2026-08-30","source":"https://www.together.ai/pricing","note":"Flat rate — input and output cost the same."},{"id":"together-llama-3.1-70b","name":"Llama 3.1 70B","tier":"mid","context_window":128000,"max_output":null,"input":0.88,"output":0.88,"cache_write":null,"cache_read":null,"modalities":["text"],"released":"2024-07","source":"https://www.together.ai/models/llama-3-1-70b","note":"Flat rate — input and output cost the same.","verified":"2026-08-29"}]},{"id":"cerebras","name":"Cerebras","pricing_url":"https://www.cerebras.ai/pricing","verification":"official_page","models":[{"id":"cerebras-gpt-oss-120b","name":"GPT OSS 120B","tier":"mid","context_window":null,"max_output":null,"input":0.35,"output":0.75,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://www.cerebras.ai/pricing","note":"Developer Tier pricing, listed at ~3000 tokens/s. Preview-tier models on Cerebras are for evaluation only, not production."},{"id":"cerebras-gemma-4-31b","name":"Gemma 4 31B","tier":"small","context_window":null,"max_output":null,"input":0.99,"output":1.49,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://www.cerebras.ai/pricing","note":"Developer Tier pricing, listed at ~1,800 tokens/s. Preview-tier models on Cerebras are for evaluation only, not production."}]},{"id":"sambanova","name":"SambaNova","pricing_url":"https://cloud.sambanova.ai/pricing","verification":"official_api","models":[{"id":"sambanova-gpt-oss-120b","name":"gpt-oss-120b","tier":"mid","context_window":null,"max_output":null,"input":0.22,"output":0.59,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://cloud.sambanova.ai/pricing","note":"SambaNova lists gpt-oss-120b at $0.22 / $0.59 per 1M tokens on its own pricing page."},{"id":"sambanova-llama-4-maverick-17b-128e-instruct","name":"Llama-4-Maverick-17B-128E-Instruct","tier":"flagship","context_window":null,"max_output":null,"input":0.63,"output":1.8,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://cloud.sambanova.ai/api/pricing","note":"SambaNova Cloud pricing API returns prices in cents per 1M tokens (63/180), converted here to USD per 1M tokens."},{"id":"sambanova-llama-3-3-70b","name":"Meta-Llama-3.3-70B-Instruct","tier":"mid","context_window":null,"max_output":null,"input":0.6,"output":1.2,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://cloud.sambanova.ai/pricing","note":"SambaNova lists Llama 3.3 70B Instruct at $0.60 / $1.20 per 1M tokens on its own pricing page."},{"id":"sambanova-qwen3-235b-a22b-instruct-2507","name":"Qwen3-235B-A22B-Instruct-2507","tier":"flagship","context_window":null,"max_output":null,"input":0.4,"output":0.8,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://cloud.sambanova.ai/api/pricing","note":"SambaNova Cloud pricing API returns prices in cents per 1M tokens (40/80), converted here to USD per 1M tokens."},{"id":"sambanova-qwen3-32b","name":"Qwen3-32B","tier":"mid","context_window":null,"max_output":null,"input":0.4,"output":0.8,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://cloud.sambanova.ai/api/pricing","note":"SambaNova Cloud pricing API returns prices in cents per 1M tokens (40/80), converted here to USD per 1M tokens."},{"id":"sambanova-deepseek-v3-2","name":"DeepSeek-V3.2","tier":"flagship","context_window":null,"max_output":null,"input":3,"output":4.5,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://cloud.sambanova.ai/pricing","note":"SambaNova lists DeepSeek V3.2 at $3.00 / $4.50 per 1M tokens on its own pricing page."},{"id":"sambanova-deepseek-r1-0528","name":"DeepSeek-R1-0528","tier":"reasoning","context_window":null,"max_output":null,"input":5,"output":7,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://cloud.sambanova.ai/api/pricing","note":"SambaNova Cloud pricing API returns prices in cents per 1M tokens (500/700), converted here to USD per 1M tokens."},{"id":"sambanova-deepseek-r1-distill-llama-70b","name":"DeepSeek-R1-Distill-Llama-70B","tier":"reasoning","context_window":null,"max_output":null,"input":0.7,"output":1.4,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://cloud.sambanova.ai/api/pricing","note":"SambaNova Cloud pricing API returns prices in cents per 1M tokens (70/140), converted here to USD per 1M tokens."},{"id":"sambanova-minimax-m2-7","name":"MiniMax-M2.7","tier":"mid","context_window":null,"max_output":null,"input":0.6,"output":2.4,"cache_write":null,"cache_read":0.06,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://cloud.sambanova.ai/pricing","note":"SambaNova lists MiniMax-M2.7 at $0.60 / $2.40 per 1M tokens, with cached input tokens at $0.06 per 1M."},{"id":"sambanova-gemma-4-31b","name":"gemma-4-31B-it","tier":"small","context_window":null,"max_output":null,"input":0.38,"output":1.15,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://cloud.sambanova.ai/pricing","note":"SambaNova lists Gemma 4 31B IT at $0.38 / $1.15 per 1M tokens on its own pricing page."},{"id":"sambanova-minimax-m3","name":"MiniMax-M3","tier":"mid","context_window":null,"max_output":null,"input":0.6,"output":2.4,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://cloud.sambanova.ai/pricing","note":"SambaNova lists MiniMax-M3 at $0.60 / $2.40 per 1M tokens on its own pricing page."}]},{"id":"perplexity","name":"Perplexity","pricing_url":"https://docs.perplexity.ai/getting-started/pricing","verification":"official_page","models":[{"id":"sonar","name":"Sonar","tier":"small","context_window":null,"max_output":null,"input":1,"output":1,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://docs.perplexity.ai/getting-started/pricing","note":"Token prices exclude the per-request search fee, billed per 1000 requests based on search context size: $5 (low), $8 (medium), $12 (high)."},{"id":"sonar-pro","name":"Sonar Pro","tier":"flagship","context_window":null,"max_output":null,"input":3,"output":15,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://docs.perplexity.ai/getting-started/pricing","note":"Token prices exclude the per-request search fee, billed per 1000 requests based on search context size: $6 (low), $10 (medium), $14 (high)."},{"id":"sonar-reasoning-pro","name":"Sonar Reasoning Pro","tier":"reasoning","context_window":null,"max_output":null,"input":2,"output":8,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://docs.perplexity.ai/getting-started/pricing","note":"Token prices exclude the per-request search fee, billed per 1000 requests based on search context size: $6 (low), $10 (medium), $14 (high)."},{"id":"sonar-deep-research","name":"Sonar Deep Research","tier":"reasoning","context_window":null,"max_output":null,"input":2,"output":8,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://docs.perplexity.ai/getting-started/pricing","note":"Additionally bills citation tokens at $2 per 1M tokens, reasoning tokens at $3 per 1M tokens, and search queries at $5 per 1000 queries. Not listed in the per-request search-context-size fee table (Sonar/Sonar Pro/Sonar Reasoning Pro only)."}]},{"id":"ai21","name":"AI21 Labs","pricing_url":"https://www.ai21.com/pricing/","verification":"official_page","models":[{"id":"jamba-mini","name":"Jamba Mini","tier":"small","context_window":null,"max_output":null,"input":0.2,"output":0.4,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://www.ai21.com/pricing/","note":"Listed as 'Efficient & lightweight model for a wide range of tasks' on AI21's official pricing page."},{"id":"jamba-large","name":"Jamba Large","tier":"flagship","context_window":null,"max_output":null,"input":2,"output":8,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://www.ai21.com/pricing/","note":"Listed as 'The most powerful and efficient long context model' on AI21's official pricing page. AI21 notes its tokenizer averages ~1 word or 6 characters per token, up to 30% more text per token than other providers' tokenizers."}]},{"id":"cloudflare","name":"Cloudflare Workers AI","pricing_url":"https://developers.cloudflare.com/workers-ai/platform/pricing/","verification":"official_page","models":[{"id":"cloudflare-llama-3.3-70b-instruct-fp8-fast","name":"Llama 3.3 70B Instruct (fp8-fast)","tier":"flagship","context_window":null,"max_output":null,"input":0.293,"output":2.253,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://developers.cloudflare.com/workers-ai/platform/pricing/","note":"Page lists this model (@cf/meta/llama-3.3-70b-instruct-fp8-fast) at both an explicit USD/1M-token rate and an equivalent neuron rate (26668 neurons per M input tokens, 204805 neurons per M output tokens at $0.011/1000 neurons); the USD figures shown here are the page's stated USD/1M figures directly."},{"id":"cloudflare-llama-4-scout-17b-16e-instruct","name":"Llama 4 Scout 17B 16E Instruct","tier":"mid","context_window":null,"max_output":null,"input":0.27,"output":0.85,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://developers.cloudflare.com/workers-ai/platform/pricing/","note":"Model id @cf/meta/llama-4-scout-17b-16e-instruct. Neuron-equivalent: 24545 neurons per M input tokens, 77273 neurons per M output tokens."},{"id":"cloudflare-llama-3.1-8b-instruct-fp8-fast","name":"Llama 3.1 8B Instruct (fp8-fast)","tier":"small","context_window":null,"max_output":null,"input":0.045,"output":0.384,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://developers.cloudflare.com/workers-ai/platform/pricing/","note":"Model id @cf/meta/llama-3.1-8b-instruct-fp8-fast. Neuron-equivalent: 4119 neurons per M input tokens, 34868 neurons per M output tokens."},{"id":"cloudflare-deepseek-v4-pro-0813","name":"DeepSeek V4 Pro (0813)","tier":"flagship","context_window":null,"max_output":null,"input":1.32,"output":3.96,"cache_write":null,"cache_read":0.044,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://developers.cloudflare.com/workers-ai/platform/pricing/","note":"Model id @cf/deepseek-ai/deepseek-v4-pro-0813. Requires a paid Workers billing method or prepaid AI Gateway credits per the page's note. Neuron-equivalent: 120000 neurons per M input, 4000 per M cached input, 360000 per M output."},{"id":"cloudflare-deepseek-v4-flash-0731","name":"DeepSeek V4 Flash (0731)","tier":"mid","context_window":null,"max_output":null,"input":0.44,"output":1.32,"cache_write":null,"cache_read":0.014,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://developers.cloudflare.com/workers-ai/platform/pricing/","note":"Model id @cf/deepseek-ai/deepseek-v4-flash-0731. Requires a paid Workers billing method or prepaid AI Gateway credits per the page's note. Neuron-equivalent: 40000 neurons per M input, 1273 per M cached input, 120000 per M output."},{"id":"cloudflare-deepseek-r1-distill-qwen-32b","name":"DeepSeek R1 Distill Qwen 32B","tier":"reasoning","context_window":null,"max_output":null,"input":0.497,"output":4.881,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://developers.cloudflare.com/workers-ai/platform/pricing/","note":"Model id @cf/deepseek-ai/deepseek-r1-distill-qwen-32b. Neuron-equivalent: 45170 neurons per M input tokens, 443756 neurons per M output tokens."},{"id":"cloudflare-qwq-32b","name":"QwQ 32B","tier":"reasoning","context_window":null,"max_output":null,"input":0.66,"output":1,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://developers.cloudflare.com/workers-ai/platform/pricing/","note":"Model id @cf/qwen/qwq-32b. Neuron-equivalent: 60000 neurons per M input tokens, 90909 neurons per M output tokens."},{"id":"cloudflare-qwen2.5-coder-32b-instruct","name":"Qwen2.5 Coder 32B Instruct","tier":"mid","context_window":null,"max_output":null,"input":0.66,"output":1,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://developers.cloudflare.com/workers-ai/platform/pricing/","note":"Model id @cf/qwen/qwen2.5-coder-32b-instruct. Neuron-equivalent: 60000 neurons per M input tokens, 90909 neurons per M output tokens."},{"id":"cloudflare-mistral-small-3.1-24b-instruct","name":"Mistral Small 3.1 24B Instruct","tier":"mid","context_window":null,"max_output":null,"input":0.351,"output":0.555,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://developers.cloudflare.com/workers-ai/platform/pricing/","note":"Model id @cf/mistralai/mistral-small-3.1-24b-instruct. Neuron-equivalent: 31876 neurons per M input tokens, 50488 neurons per M output tokens."},{"id":"cloudflare-gpt-oss-120b","name":"GPT-OSS 120B","tier":"flagship","context_window":null,"max_output":null,"input":0.35,"output":0.75,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://developers.cloudflare.com/workers-ai/platform/pricing/","note":"Model id @cf/openai/gpt-oss-120b. Neuron-equivalent: 31818 neurons per M input tokens, 68182 neurons per M output tokens."},{"id":"cloudflare-gpt-oss-20b","name":"GPT-OSS 20B","tier":"small","context_window":null,"max_output":null,"input":0.2,"output":0.3,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://developers.cloudflare.com/workers-ai/platform/pricing/","note":"Model id @cf/openai/gpt-oss-20b. Neuron-equivalent: 18182 neurons per M input tokens, 27273 neurons per M output tokens."}]},{"id":"novita","name":"Novita AI","pricing_url":"https://novita.ai/pricing","verification":"official_page","models":[{"id":"novita-deepseek-v3.1-terminus","name":"DeepSeek V3.1 Terminus","tier":"flagship","context_window":128000,"max_output":null,"input":0.27,"output":1,"cache_write":null,"cache_read":0.135,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://novita.ai/pricing","note":"Serverless per-token pricing on Novita's model catalog; dedicated GPU deployments are billed per hour instead."},{"id":"novita-deepseek-r1-0528","name":"DeepSeek R1 0528","tier":"reasoning","context_window":160000,"max_output":null,"input":0.7,"output":2.5,"cache_write":null,"cache_read":0.35,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://novita.ai/pricing","note":"Serverless per-token pricing on Novita's model catalog; dedicated GPU deployments are billed per hour instead."},{"id":"novita-qwen3-235b-a22b-instruct-2507","name":"Qwen3 235B A22B Instruct 2507","tier":"flagship","context_window":128000,"max_output":null,"input":0.09,"output":0.58,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://novita.ai/pricing","note":"Serverless per-token pricing on Novita's model catalog; dedicated GPU deployments are billed per hour instead."},{"id":"novita-qwen3-coder-480b-a35b-instruct","name":"Qwen3 Coder 480B A35B Instruct","tier":"flagship","context_window":256000,"max_output":null,"input":0.38,"output":1.55,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://novita.ai/pricing","note":"Serverless per-token pricing on Novita's model catalog; dedicated GPU deployments are billed per hour instead."},{"id":"novita-kimi-k2-instruct","name":"Kimi K2 Instruct","tier":"flagship","context_window":128000,"max_output":null,"input":0.57,"output":2.3,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://novita.ai/pricing","note":"Serverless per-token pricing on Novita's model catalog; dedicated GPU deployments are billed per hour instead."},{"id":"novita-kimi-k2-thinking","name":"Kimi K2 Thinking","tier":"reasoning","context_window":256000,"max_output":null,"input":0.6,"output":2.5,"cache_write":null,"cache_read":0.15,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://novita.ai/pricing","note":"Serverless per-token pricing on Novita's model catalog; dedicated GPU deployments are billed per hour instead."},{"id":"novita-glm-4.6","name":"GLM 4.6","tier":"flagship","context_window":200000,"max_output":null,"input":0.55,"output":2.2,"cache_write":null,"cache_read":0.11,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://novita.ai/pricing","note":"Serverless per-token pricing on Novita's model catalog; dedicated GPU deployments are billed per hour instead."},{"id":"novita-llama-4-maverick-instruct","name":"Llama 4 Maverick Instruct","tier":"flagship","context_window":1000000,"max_output":null,"input":0.27,"output":0.85,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://novita.ai/pricing","note":"Serverless per-token pricing on Novita's model catalog; dedicated GPU deployments are billed per hour instead."},{"id":"novita-llama-4-scout-instruct","name":"Llama 4 Scout Instruct","tier":"mid","context_window":128000,"max_output":null,"input":0.18,"output":0.59,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://novita.ai/pricing","note":"Serverless per-token pricing on Novita's model catalog; dedicated GPU deployments are billed per hour instead."},{"id":"novita-gpt-oss-120b","name":"GPT-OSS 120B","tier":"mid","context_window":128000,"max_output":null,"input":0.05,"output":0.25,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://novita.ai/pricing","note":"Serverless per-token pricing on Novita's model catalog; dedicated GPU deployments are billed per hour instead."},{"id":"novita-gpt-oss-20b","name":"GPT-OSS 20B","tier":"small","context_window":128000,"max_output":null,"input":0.04,"output":0.15,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://novita.ai/pricing","note":"Serverless per-token pricing on Novita's model catalog; dedicated GPU deployments are billed per hour instead."},{"id":"novita-mistral-nemo","name":"Mistral Nemo","tier":"small","context_window":59000,"max_output":null,"input":0.04,"output":0.17,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://novita.ai/pricing","note":"Serverless per-token pricing on Novita's model catalog; dedicated GPU deployments are billed per hour instead."}]},{"id":"baseten","name":"Baseten","pricing_url":"https://www.baseten.co/pricing/","verification":"official_page","models":[{"id":"baseten-glm-5.3","name":"GLM-5.3","tier":"flagship","context_window":null,"max_output":null,"input":1.4,"output":4.4,"cache_write":null,"cache_read":0.14,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://www.baseten.co/pricing/","note":"Baseten Model APIs price, listed per 1M tokens; separate from Baseten's per-minute Dedicated Deployment GPU pricing."},{"id":"baseten-glm-5.3-flash","name":"GLM-5.3-Flash","tier":"mid","context_window":null,"max_output":null,"input":0.15,"output":0.5,"cache_write":null,"cache_read":0.03,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://www.baseten.co/pricing/","note":"Baseten Model APIs price, listed per 1M tokens; separate from Baseten's per-minute Dedicated Deployment GPU pricing."},{"id":"baseten-glm-5.2","name":"GLM-5.2","tier":"flagship","context_window":null,"max_output":null,"input":1.4,"output":4.4,"cache_write":null,"cache_read":0.14,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://www.baseten.co/pricing/","note":"Baseten Model APIs price, listed per 1M tokens; separate from Baseten's per-minute Dedicated Deployment GPU pricing."},{"id":"baseten-glm-5.2-fast","name":"GLM-5.2 Fast","tier":"flagship","context_window":null,"max_output":null,"input":2.1,"output":6.6,"cache_write":null,"cache_read":0.21,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://www.baseten.co/pricing/","note":"Baseten Model APIs price, listed per 1M tokens, for the latency-optimized 'Fast' variant of GLM-5.2; separate from Baseten's per-minute Dedicated Deployment GPU pricing."},{"id":"baseten-glm-4.7","name":"GLM 4.7","tier":"mid","context_window":null,"max_output":null,"input":0.6,"output":2.2,"cache_write":null,"cache_read":0.12,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://www.baseten.co/pricing/","note":"Baseten Model APIs price, listed per 1M tokens; separate from Baseten's per-minute Dedicated Deployment GPU pricing."},{"id":"baseten-deepseek-v4-pro-0813","name":"DeepSeek V4 Pro 0813","tier":"flagship","context_window":null,"max_output":null,"input":1.32,"output":3.96,"cache_write":null,"cache_read":0.132,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://www.baseten.co/pricing/","note":"Baseten Model APIs price, listed per 1M tokens; separate from Baseten's per-minute Dedicated Deployment GPU pricing."},{"id":"baseten-deepseek-v4-pro","name":"DeepSeek V4 Pro","tier":"flagship","context_window":null,"max_output":null,"input":1.74,"output":3.48,"cache_write":null,"cache_read":0.145,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://www.baseten.co/pricing/","note":"Baseten Model APIs price, listed per 1M tokens; separate from Baseten's per-minute Dedicated Deployment GPU pricing."},{"id":"baseten-deepseek-v4-flash-0731","name":"DeepSeek V4 Flash 0731","tier":"mid","context_window":null,"max_output":null,"input":0.13,"output":0.26,"cache_write":null,"cache_read":0.028,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://www.baseten.co/pricing/","note":"Baseten Model APIs price, listed per 1M tokens; separate from Baseten's per-minute Dedicated Deployment GPU pricing."},{"id":"baseten-kimi-k3","name":"Kimi K3","tier":"flagship","context_window":null,"max_output":null,"input":3,"output":15,"cache_write":null,"cache_read":0.3,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://www.baseten.co/pricing/","note":"Baseten Model APIs price, listed per 1M tokens; separate from Baseten's per-minute Dedicated Deployment GPU pricing."},{"id":"baseten-kimi-k2.6","name":"Kimi K2.6","tier":"flagship","context_window":null,"max_output":null,"input":0.95,"output":4,"cache_write":null,"cache_read":0.16,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://www.baseten.co/pricing/","note":"Baseten Model APIs price, listed per 1M tokens; separate from Baseten's per-minute Dedicated Deployment GPU pricing."},{"id":"baseten-kimi-k2.7-code","name":"Kimi K2.7 Code","tier":"flagship","context_window":null,"max_output":null,"input":0.95,"output":4,"cache_write":null,"cache_read":0.16,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://www.baseten.co/pricing/","note":"Baseten Model APIs price, listed per 1M tokens, for the coding-tuned Kimi K2.7 Code variant; separate from Baseten's per-minute Dedicated Deployment GPU pricing."},{"id":"baseten-gpt-oss-120b","name":"GPT OSS 120B","tier":"mid","context_window":null,"max_output":null,"input":0.1,"output":0.5,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://www.baseten.co/pricing/","note":"Baseten Model APIs price, listed per 1M tokens; the page lists no cache-input price for this model. Separate from Baseten's per-minute Dedicated Deployment GPU pricing."}]},{"id":"azure-openai","name":"Azure OpenAI Service","pricing_url":"https://azure.microsoft.com/en-us/pricing/details/cognitive-services/openai-service/","verification":"official_page","models":[{"id":"azure-openai-gpt-chat-latest","name":"GPT-Chat Latest 05052026","tier":"flagship","context_window":null,"max_output":null,"input":5,"output":30,"cache_write":null,"cache_read":0.5,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://azure.microsoft.com/en-us/pricing/details/cognitive-services/openai-service/","note":"Global Standard (Pay-As-You-Go) pricing; table is quoted directly in USD per 1M tokens, no conversion needed."},{"id":"azure-openai-gpt-5-6-sol-short-context","name":"GPT-5.6-sol (short context)","tier":"flagship","context_window":null,"max_output":null,"input":5,"output":30,"cache_write":6.25,"cache_read":0.5,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://azure.microsoft.com/en-us/pricing/details/cognitive-services/openai-service/","note":"Global Standard tier, short-context row; price table is in USD per 1M tokens already, no conversion needed."},{"id":"azure-openai-gpt-5-6-sol-long-context","name":"GPT-5.6-sol (long context)","tier":"flagship","context_window":null,"max_output":null,"input":10,"output":45,"cache_write":12.5,"cache_read":1,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://azure.microsoft.com/en-us/pricing/details/cognitive-services/openai-service/","note":"Global Standard tier, long-context row (higher per-token rate applies beyond the model's short-context threshold); USD per 1M tokens as published."},{"id":"azure-openai-gpt-5-6-terra-short-context","name":"GPT-5.6-terra (short context)","tier":"mid","context_window":null,"max_output":null,"input":2,"output":12,"cache_write":2.5,"cache_read":0.2,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://azure.microsoft.com/en-us/pricing/details/cognitive-services/openai-service/","note":"Global Standard tier, short-context row; USD per 1M tokens as published."},{"id":"azure-openai-gpt-5-6-terra-long-context","name":"GPT-5.6-terra (long context)","tier":"mid","context_window":null,"max_output":null,"input":4,"output":18,"cache_write":5,"cache_read":0.4,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://azure.microsoft.com/en-us/pricing/details/cognitive-services/openai-service/","note":"Global Standard tier, long-context row; USD per 1M tokens as published."},{"id":"azure-openai-gpt-5-6-luna-short-context","name":"GPT-5.6-luna (short context)","tier":"small","context_window":null,"max_output":null,"input":0.2,"output":1.2,"cache_write":0.25,"cache_read":0.02,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://azure.microsoft.com/en-us/pricing/details/cognitive-services/openai-service/","note":"Global Standard tier, short-context row; USD per 1M tokens as published."},{"id":"azure-openai-gpt-5-6-luna-long-context","name":"GPT-5.6-luna (long context)","tier":"small","context_window":null,"max_output":null,"input":0.4,"output":1.8,"cache_write":0.5,"cache_read":0.04,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://azure.microsoft.com/en-us/pricing/details/cognitive-services/openai-service/","note":"Global Standard tier, long-context row; USD per 1M tokens as published."},{"id":"azure-openai-gpt-5-5","name":"GPT-5.5","tier":"flagship","context_window":null,"max_output":null,"input":5,"output":30,"cache_write":null,"cache_read":0.5,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://azure.microsoft.com/en-us/pricing/details/cognitive-services/openai-service/","note":"Global Standard tier; USD per 1M tokens as published."},{"id":"azure-openai-gpt-5-5-long-context","name":"GPT-5.5 Long Context","tier":"flagship","context_window":null,"max_output":null,"input":10,"output":45,"cache_write":null,"cache_read":1,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://azure.microsoft.com/en-us/pricing/details/cognitive-services/openai-service/","note":"Global Standard tier, long-context row; USD per 1M tokens as published."},{"id":"azure-openai-gpt-5-4-short-context","name":"GPT-5.4 (<272k context length)","tier":"flagship","context_window":null,"max_output":null,"input":2.5,"output":15,"cache_write":null,"cache_read":0.25,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://azure.microsoft.com/en-us/pricing/details/cognitive-services/openai-service/","note":"Global Standard tier, under-272k-context row; USD per 1M tokens as published."},{"id":"azure-openai-gpt-5-4-long-context","name":"GPT-5.4 (>272k context length)","tier":"flagship","context_window":null,"max_output":null,"input":5,"output":22.5,"cache_write":null,"cache_read":0.5,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://azure.microsoft.com/en-us/pricing/details/cognitive-services/openai-service/","note":"Global Standard tier, over-272k-context row; USD per 1M tokens as published."},{"id":"azure-openai-gpt-5-4-pro-short-context","name":"GPT-5.4 Pro (<272k context length)","tier":"flagship","context_window":null,"max_output":null,"input":30,"output":180,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://azure.microsoft.com/en-us/pricing/details/cognitive-services/openai-service/","note":"Global Standard tier, under-272k-context row; no cached-input rate is published for this model; USD per 1M tokens as published."},{"id":"azure-openai-gpt-5-4-pro-long-context","name":"GPT-5.4 Pro (>272k context length)","tier":"flagship","context_window":null,"max_output":null,"input":60,"output":270,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://azure.microsoft.com/en-us/pricing/details/cognitive-services/openai-service/","note":"Global Standard tier, over-272k-context row; no cached-input rate is published for this model; USD per 1M tokens as published."},{"id":"azure-openai-gpt-5-4-mini","name":"GPT-5.4 mini","tier":"mid","context_window":null,"max_output":null,"input":0.75,"output":4.5,"cache_write":null,"cache_read":0.075,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://azure.microsoft.com/en-us/pricing/details/cognitive-services/openai-service/","note":"Global Standard tier; USD per 1M tokens as published."},{"id":"azure-openai-gpt-5-4-nano","name":"GPT-5.4 nano","tier":"small","context_window":null,"max_output":null,"input":0.2,"output":1.25,"cache_write":null,"cache_read":0.02,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://azure.microsoft.com/en-us/pricing/details/cognitive-services/openai-service/","note":"Global Standard tier; USD per 1M tokens as published."},{"id":"azure-openai-gpt-5-3-codex","name":"GPT-5.3 Codex","tier":"reasoning","context_window":null,"max_output":null,"input":1.75,"output":14,"cache_write":null,"cache_read":0.175,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://azure.microsoft.com/en-us/pricing/details/cognitive-services/openai-service/","note":"Global Standard tier; USD per 1M tokens as published."},{"id":"azure-openai-gpt-5-3-chat","name":"GPT-5.3 Chat","tier":"flagship","context_window":null,"max_output":null,"input":1.75,"output":14,"cache_write":null,"cache_read":0.175,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://azure.microsoft.com/en-us/pricing/details/cognitive-services/openai-service/","note":"Global Standard tier; USD per 1M tokens as published."},{"id":"azure-openai-gpt-5-2","name":"GPT-5.2","tier":"flagship","context_window":null,"max_output":null,"input":1.75,"output":14,"cache_write":null,"cache_read":0.175,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://azure.microsoft.com/en-us/pricing/details/cognitive-services/openai-service/","note":"Global Standard tier; USD per 1M tokens as published."},{"id":"azure-openai-gpt-5-2-codex","name":"GPT-5.2 Codex","tier":"reasoning","context_window":null,"max_output":null,"input":1.75,"output":14,"cache_write":null,"cache_read":0.175,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://azure.microsoft.com/en-us/pricing/details/cognitive-services/openai-service/","note":"Global Standard tier; USD per 1M tokens as published."},{"id":"azure-openai-gpt-5-2-chat-latest","name":"GPT-5.2-chat latest","tier":"flagship","context_window":null,"max_output":null,"input":1.75,"output":14,"cache_write":null,"cache_read":0.175,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://azure.microsoft.com/en-us/pricing/details/cognitive-services/openai-service/","note":"Global Standard tier; USD per 1M tokens as published."},{"id":"azure-openai-gpt-5-1","name":"GPT-5.1","tier":"flagship","context_window":null,"max_output":null,"input":1.25,"output":10,"cache_write":null,"cache_read":0.125,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://azure.microsoft.com/en-us/pricing/details/cognitive-services/openai-service/","note":"Global Standard tier; USD per 1M tokens as published."},{"id":"azure-openai-gpt-5-1-chat","name":"GPT-5.1-chat","tier":"flagship","context_window":null,"max_output":null,"input":1.25,"output":10,"cache_write":null,"cache_read":0.125,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://azure.microsoft.com/en-us/pricing/details/cognitive-services/openai-service/","note":"Global Standard tier; USD per 1M tokens as published."},{"id":"azure-openai-gpt-5-1-codex","name":"GPT-5.1-codex","tier":"reasoning","context_window":null,"max_output":null,"input":1.25,"output":10,"cache_write":null,"cache_read":0.125,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://azure.microsoft.com/en-us/pricing/details/cognitive-services/openai-service/","note":"Global Standard tier; USD per 1M tokens as published."},{"id":"azure-openai-gpt-5-1-codex-max","name":"GPT-5.1-codex-max","tier":"reasoning","context_window":null,"max_output":null,"input":1.25,"output":10,"cache_write":null,"cache_read":0.125,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://azure.microsoft.com/en-us/pricing/details/cognitive-services/openai-service/","note":"Global Standard tier; USD per 1M tokens as published."},{"id":"azure-openai-gpt-5-1-codex-mini","name":"GPT-5.1-codex-mini","tier":"reasoning","context_window":null,"max_output":null,"input":0.25,"output":2,"cache_write":null,"cache_read":0.025,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://azure.microsoft.com/en-us/pricing/details/cognitive-services/openai-service/","note":"Global Standard tier; USD per 1M tokens as published."},{"id":"azure-openai-gpt-5","name":"GPT-5 (2025-08-07)","tier":"flagship","context_window":null,"max_output":null,"input":1.25,"output":10,"cache_write":null,"cache_read":0.125,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://azure.microsoft.com/en-us/pricing/details/cognitive-services/openai-service/","note":"Global Standard tier; USD per 1M tokens as published."},{"id":"azure-openai-gpt-5-pro","name":"GPT-5 Pro","tier":"flagship","context_window":null,"max_output":null,"input":15,"output":120,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://azure.microsoft.com/en-us/pricing/details/cognitive-services/openai-service/","note":"Global Standard tier; no cached-input rate is published for this model; USD per 1M tokens as published."},{"id":"azure-openai-gpt-5-codex","name":"GPT-5 Codex","tier":"reasoning","context_window":null,"max_output":null,"input":1.25,"output":10,"cache_write":null,"cache_read":0.125,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://azure.microsoft.com/en-us/pricing/details/cognitive-services/openai-service/","note":"Global Standard tier; USD per 1M tokens as published."},{"id":"azure-openai-gpt-5-mini","name":"GPT-5-mini","tier":"mid","context_window":null,"max_output":null,"input":0.25,"output":2,"cache_write":null,"cache_read":0.025,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://azure.microsoft.com/en-us/pricing/details/cognitive-services/openai-service/","note":"Global Standard tier; USD per 1M tokens as published."},{"id":"azure-openai-gpt-5-nano","name":"GPT-5-nano","tier":"small","context_window":null,"max_output":null,"input":0.05,"output":0.4,"cache_write":null,"cache_read":0.005,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://azure.microsoft.com/en-us/pricing/details/cognitive-services/openai-service/","note":"Global Standard tier; USD per 1M tokens as published."},{"id":"azure-openai-gpt-5-chat","name":"GPT-5 chat","tier":"flagship","context_window":null,"max_output":null,"input":1.25,"output":10,"cache_write":null,"cache_read":0.125,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://azure.microsoft.com/en-us/pricing/details/cognitive-services/openai-service/","note":"Global Standard tier; USD per 1M tokens as published."},{"id":"azure-openai-o3-deep-research","name":"o3-deep research","tier":"reasoning","context_window":null,"max_output":null,"input":10,"output":40,"cache_write":null,"cache_read":2.5,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://azure.microsoft.com/en-us/pricing/details/cognitive-services/openai-service/","note":"Global Standard tier; USD per 1M tokens as published. Deep Research also incurs separate Grounding-with-Bing-Search and base-model charges per Azure's page, not reflected in this per-token figure."},{"id":"azure-openai-o3","name":"o3 (2025-04-16)","tier":"reasoning","context_window":200000,"max_output":null,"input":2,"output":8,"cache_write":null,"cache_read":0.5,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://azure.microsoft.com/en-us/pricing/details/cognitive-services/openai-service/","note":"Global Standard tier; USD per 1M tokens as published; Azure states o3 has a 200K token context window."},{"id":"azure-openai-o4-mini","name":"o4-mini (2025-04-16)","tier":"reasoning","context_window":200000,"max_output":null,"input":1.1,"output":4.4,"cache_write":null,"cache_read":0.275,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://azure.microsoft.com/en-us/pricing/details/cognitive-services/openai-service/","note":"Global Standard tier; USD per 1M tokens as published; Azure states o4-mini has a 200K token context window."},{"id":"azure-openai-gpt-4-1","name":"GPT-4.1-2025-04-14","tier":"flagship","context_window":1000000,"max_output":null,"input":2,"output":8,"cache_write":null,"cache_read":0.5,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://azure.microsoft.com/en-us/pricing/details/cognitive-services/openai-service/","note":"Global Standard tier; USD per 1M tokens as published; Azure states the GPT-4.1 series has a 1 million token context window."},{"id":"azure-openai-gpt-4-1-mini","name":"GPT-4.1-mini-2025-04-14","tier":"mid","context_window":1000000,"max_output":null,"input":0.4,"output":1.6,"cache_write":null,"cache_read":0.1,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://azure.microsoft.com/en-us/pricing/details/cognitive-services/openai-service/","note":"Global Standard tier; USD per 1M tokens as published; Azure states the GPT-4.1 series has a 1 million token context window."},{"id":"azure-openai-gpt-4-1-nano","name":"GPT-4.1-nano-2025-04-14","tier":"small","context_window":1000000,"max_output":null,"input":0.1,"output":0.4,"cache_write":null,"cache_read":0.025,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://azure.microsoft.com/en-us/pricing/details/cognitive-services/openai-service/","note":"Global Standard tier; USD per 1M tokens as published; Azure states the GPT-4.1 series has a 1 million token context window."},{"id":"azure-openai-gpt-4-5-preview","name":"GPT-4.5-Preview-2025-02-27","tier":"flagship","context_window":128000,"max_output":null,"input":75,"output":150,"cache_write":null,"cache_read":37.5,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://azure.microsoft.com/en-us/pricing/details/cognitive-services/openai-service/","note":"Global Standard tier; USD per 1M tokens as published; Azure states GPT-4.5-preview has 128K context."},{"id":"azure-openai-o1","name":"o1 (2024-12-17)","tier":"reasoning","context_window":200000,"max_output":null,"input":15,"output":60,"cache_write":null,"cache_read":7.5,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://azure.microsoft.com/en-us/pricing/details/cognitive-services/openai-service/","note":"Global Standard tier; USD per 1M tokens as published; Azure states o1 has 200K context."},{"id":"azure-openai-o1-preview","name":"o1 preview (2024-09-12)","tier":"reasoning","context_window":null,"max_output":null,"input":15,"output":60,"cache_write":null,"cache_read":7.5,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://azure.microsoft.com/en-us/pricing/details/cognitive-services/openai-service/","note":"Global Standard tier; USD per 1M tokens as published."},{"id":"azure-openai-o3-mini","name":"o3 mini (2025-01-31)","tier":"reasoning","context_window":200000,"max_output":100000,"input":1.1,"output":4.4,"cache_write":null,"cache_read":0.55,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://azure.microsoft.com/en-us/pricing/details/cognitive-services/openai-service/","note":"Global Standard tier; USD per 1M tokens as published; Azure states o3-mini has a 200K token context input window and a 100K token maximum output."},{"id":"azure-openai-o1-mini","name":"o1-mini (2024-09-12)","tier":"reasoning","context_window":128000,"max_output":null,"input":1.1,"output":4.4,"cache_write":null,"cache_read":0.55,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://azure.microsoft.com/en-us/pricing/details/cognitive-services/openai-service/","note":"Global Standard tier; USD per 1M tokens as published; Azure states o1-mini has 128K context input."},{"id":"azure-openai-gpt-4o-2024-11-20","name":"GPT-4o-2024-1120","tier":"flagship","context_window":128000,"max_output":null,"input":2.5,"output":10,"cache_write":null,"cache_read":1.25,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://azure.microsoft.com/en-us/pricing/details/cognitive-services/openai-service/","note":"Global Standard tier; USD per 1M tokens as published; Azure states GPT-4o has 128K context."},{"id":"azure-openai-gpt-4o-2024-08-06","name":"GPT-4o-2024-08-06","tier":"flagship","context_window":128000,"max_output":null,"input":2.5,"output":10,"cache_write":null,"cache_read":1.25,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://azure.microsoft.com/en-us/pricing/details/cognitive-services/openai-service/","note":"Global Standard tier; USD per 1M tokens as published; Azure states GPT-4o has 128K context."},{"id":"azure-openai-gpt-4o-2024-05-13","name":"GPT-4o-2024-0513","tier":"flagship","context_window":128000,"max_output":null,"input":5,"output":15,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://azure.microsoft.com/en-us/pricing/details/cognitive-services/openai-service/","note":"Global Standard tier; no cached-input rate is published for this earlier snapshot; USD per 1M tokens as published; Azure states GPT-4o has 128K context."},{"id":"azure-openai-gpt-4o-mini","name":"GPT-4o-mini-0718","tier":"small","context_window":128000,"max_output":null,"input":0.15,"output":0.6,"cache_write":null,"cache_read":0.075,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://azure.microsoft.com/en-us/pricing/details/cognitive-services/openai-service/","note":"Global Standard tier; USD per 1M tokens as published; Azure states GPT-4o mini has 128K context."}]},{"id":"replicate","name":"Replicate","pricing_url":"https://replicate.com/pricing","verification":"official_page","models":[{"id":"replicate-deepseek-ai-deepseek-r1","name":"deepseek-ai/deepseek-r1","tier":"reasoning","context_window":null,"max_output":null,"input":3.75,"output":10,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://replicate.com/deepseek-ai/deepseek-r1","note":"Replicate bills output at $0.01 per thousand tokens ($10 per 1M, converted) and input at $3.75 per million tokens."},{"id":"replicate-anthropic-claude-3-5-haiku","name":"anthropic/claude-3.5-haiku","tier":"small","context_window":null,"max_output":null,"input":1,"output":5,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://replicate.com/anthropic/claude-3.5-haiku","note":"Replicate quotes $1 per million input tokens and $5 per million output tokens directly; no conversion needed."},{"id":"replicate-anthropic-claude-4-sonnet","name":"anthropic/claude-4-sonnet","tier":"mid","context_window":null,"max_output":null,"input":3,"output":15,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://replicate.com/anthropic/claude-4-sonnet","note":"Replicate lists input at $3 per million tokens and output at $0.015 per thousand tokens ($15 per 1M, converted)."},{"id":"replicate-anthropic-claude-4-5-haiku","name":"anthropic/claude-4.5-haiku","tier":"small","context_window":null,"max_output":null,"input":1,"output":5,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://replicate.com/anthropic/claude-4.5-haiku","note":"Replicate quotes $1 per million input tokens and $5 per million output tokens directly; no conversion needed."},{"id":"replicate-anthropic-claude-4-5-sonnet","name":"anthropic/claude-4.5-sonnet","tier":"mid","context_window":null,"max_output":null,"input":3,"output":15,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://replicate.com/anthropic/claude-4.5-sonnet","note":"Replicate lists input at $3 per million tokens and output at $0.015 per thousand tokens ($15 per 1M, converted)."},{"id":"replicate-anthropic-claude-opus-4-6","name":"anthropic/claude-opus-4.6","tier":"flagship","context_window":null,"max_output":null,"input":5,"output":25,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://replicate.com/anthropic/claude-opus-4.6","note":"Replicate lists input at $5 per million tokens and output at $0.025 per thousand tokens ($25 per 1M, converted)."},{"id":"replicate-deepseek-ai-deepseek-v3","name":"deepseek-ai/deepseek-v3","tier":"mid","context_window":null,"max_output":null,"input":1.45,"output":1.45,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://replicate.com/deepseek-ai/deepseek-v3","note":"Replicate quotes a flat $1.45 per million tokens for both input and output; no conversion needed."},{"id":"replicate-deepseek-ai-deepseek-v3-1","name":"deepseek-ai/deepseek-v3.1","tier":"mid","context_window":null,"max_output":null,"input":0.672,"output":2.016,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://replicate.com/deepseek-ai/deepseek-v3.1","note":"Replicate quotes $0.672 per million input tokens and $2.016 per million output tokens directly; no conversion needed."},{"id":"replicate-google-gemini-2-5-flash","name":"google/gemini-2.5-flash","tier":"small","context_window":null,"max_output":null,"input":0.3,"output":2.5,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://replicate.com/google/gemini-2.5-flash","note":"Replicate quotes $0.30 per million input tokens and $2.50 per million output tokens directly; no conversion needed."},{"id":"replicate-google-gemini-3-pro","name":"google/gemini-3-pro","tier":"flagship","context_window":null,"max_output":null,"input":2,"output":12,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://replicate.com/google/gemini-3-pro","note":"Base tier for prompts up to 200,000 input tokens: Replicate lists input at $2 per million tokens and output at $0.012 per thousand tokens ($12 per 1M, converted). Above 200,000 input tokens Replicate's own tier switches to $4 per million input and $0.018 per thousand output ($18 per 1M)."},{"id":"replicate-google-gemini-3-1-pro","name":"google/gemini-3.1-pro","tier":"flagship","context_window":null,"max_output":null,"input":2,"output":12,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://replicate.com/google/gemini-3.1-pro","note":"Base tier for prompts up to 200,000 input tokens: Replicate lists input at $2 per million tokens and output at $0.012 per thousand tokens ($12 per 1M, converted). Above 200,000 input tokens Replicate's own tier switches to $4 per million input and $0.018 per thousand output ($18 per 1M)."},{"id":"replicate-ibm-granite-granite-3-3-8b-instruct","name":"ibm-granite/granite-3.3-8b-instruct","tier":"small","context_window":null,"max_output":null,"input":0.03,"output":0.25,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://replicate.com/ibm-granite/granite-3.3-8b-instruct","note":"Replicate quotes $0.03 per million input tokens and $0.25 per million output tokens directly; no conversion needed."},{"id":"replicate-meta-meta-llama-3-70b-instruct","name":"meta/meta-llama-3-70b-instruct","tier":"mid","context_window":null,"max_output":null,"input":0.65,"output":2.75,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://replicate.com/meta/meta-llama-3-70b-instruct","note":"Replicate quotes $0.65 per million input tokens and $2.75 per million output tokens directly; no conversion needed."},{"id":"replicate-meta-meta-llama-3-8b-instruct","name":"meta/meta-llama-3-8b-instruct","tier":"small","context_window":null,"max_output":null,"input":0.05,"output":0.25,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://replicate.com/meta/meta-llama-3-8b-instruct","note":"Replicate quotes $0.05 per million input tokens and $0.25 per million output tokens directly; no conversion needed."},{"id":"replicate-moonshotai-kimi-k2-5","name":"moonshotai/kimi-k2.5","tier":"mid","context_window":null,"max_output":null,"input":0.72,"output":3.6,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://replicate.com/moonshotai/kimi-k2.5","note":"Replicate quotes $0.72 per million input tokens and $3.60 per million output tokens directly; no conversion needed."},{"id":"replicate-openai-gpt-4-1","name":"openai/gpt-4.1","tier":"mid","context_window":null,"max_output":null,"input":2,"output":8,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://replicate.com/openai/gpt-4.1","note":"Replicate quotes $2 per million input tokens and $8 per million output tokens directly; no conversion needed."},{"id":"replicate-openai-gpt-4-1-mini","name":"openai/gpt-4.1-mini","tier":"small","context_window":null,"max_output":null,"input":0.4,"output":1.6,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://replicate.com/openai/gpt-4.1-mini","note":"Replicate quotes $0.40 per million input tokens and $1.60 per million output tokens directly; no conversion needed."},{"id":"replicate-openai-gpt-4-1-nano","name":"openai/gpt-4.1-nano","tier":"small","context_window":null,"max_output":null,"input":0.1,"output":0.4,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://replicate.com/openai/gpt-4.1-nano","note":"Replicate quotes $0.10 per million input tokens and $0.40 per million output tokens directly; no conversion needed."},{"id":"replicate-openai-gpt-4o","name":"openai/gpt-4o","tier":"mid","context_window":null,"max_output":null,"input":2.5,"output":10,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://replicate.com/openai/gpt-4o","note":"Replicate lists input at $2.50 per million tokens and output at $0.01 per thousand tokens ($10 per 1M, converted)."},{"id":"replicate-openai-gpt-4o-mini","name":"openai/gpt-4o-mini","tier":"small","context_window":null,"max_output":null,"input":0.15,"output":0.6,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://replicate.com/openai/gpt-4o-mini","note":"Replicate quotes $0.15 per million input tokens and $0.60 per million output tokens directly; no conversion needed."},{"id":"replicate-openai-gpt-5","name":"openai/gpt-5","tier":"flagship","context_window":null,"max_output":null,"input":1.25,"output":10,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://replicate.com/openai/gpt-5","note":"Replicate lists input at $1.25 per million tokens and output at $0.01 per thousand tokens ($10 per 1M, converted)."},{"id":"replicate-openai-gpt-5-2","name":"openai/gpt-5.2","tier":"flagship","context_window":null,"max_output":null,"input":1.75,"output":14,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://replicate.com/openai/gpt-5.2","note":"Replicate lists input at $1.75 per million tokens and output at $0.014 per thousand tokens ($14 per 1M, converted)."},{"id":"replicate-openai-gpt-5-mini","name":"openai/gpt-5-mini","tier":"mid","context_window":null,"max_output":null,"input":0.25,"output":2,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://replicate.com/openai/gpt-5-mini","note":"Replicate quotes $0.25 per million input tokens and $2 per million output tokens directly; no conversion needed."},{"id":"replicate-openai-gpt-5-nano","name":"openai/gpt-5-nano","tier":"small","context_window":null,"max_output":null,"input":0.05,"output":0.4,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://replicate.com/openai/gpt-5-nano","note":"Replicate quotes $0.05 per million input tokens and $0.40 per million output tokens directly; no conversion needed."},{"id":"replicate-openai-gpt-oss-120b","name":"openai/gpt-oss-120b","tier":"mid","context_window":null,"max_output":null,"input":0.18,"output":0.72,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://replicate.com/openai/gpt-oss-120b","note":"Replicate quotes $0.18 per million input tokens and $0.72 per million output tokens directly; no conversion needed."},{"id":"replicate-openai-gpt-oss-20b","name":"openai/gpt-oss-20b","tier":"small","context_window":null,"max_output":null,"input":0.09,"output":0.36,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://replicate.com/openai/gpt-oss-20b","note":"Replicate quotes $0.09 per million input tokens and $0.36 per million output tokens directly; no conversion needed."},{"id":"replicate-openai-o1","name":"openai/o1","tier":"reasoning","context_window":null,"max_output":null,"input":15,"output":60,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://replicate.com/openai/o1","note":"Replicate lists input at $0.015 per thousand tokens ($15 per 1M, converted) and output at $0.06 per thousand tokens ($60 per 1M, converted)."},{"id":"replicate-openai-o1-mini","name":"openai/o1-mini","tier":"reasoning","context_window":null,"max_output":null,"input":1.1,"output":4.4,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://replicate.com/openai/o1-mini","note":"Replicate quotes $1.10 per million input tokens and $4.40 per million output tokens directly; no conversion needed."},{"id":"replicate-openai-o4-mini","name":"openai/o4-mini","tier":"reasoning","context_window":null,"max_output":null,"input":1,"output":4,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://replicate.com/openai/o4-mini","note":"Replicate quotes $1 per million input tokens and $4 per million output tokens directly; no conversion needed."},{"id":"replicate-qwen-qwen3-235b-a22b-instruct-2507","name":"qwen/qwen3-235b-a22b-instruct-2507","tier":"mid","context_window":null,"max_output":null,"input":0.264,"output":1.06,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://replicate.com/qwen/qwen3-235b-a22b-instruct-2507","note":"Replicate quotes $0.264 per million input tokens and $1.06 per million output tokens directly; no conversion needed."}]},{"id":"parasail","name":"Parasail","pricing_url":"https://parasail.io/pricing","verification":"official_page","models":[{"id":"parasail-kimi-k3","name":"Kimi K3","tier":"flagship","context_window":null,"max_output":null,"input":3,"output":15,"cache_write":null,"cache_read":0.3,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://parasail.io/pricing","note":"Parasail elastic-endpoint serverless pricing, listed by the seller as price per 1M tokens: $3.0/1M input, $15.0/1M output, $0.3/1M cache read; no conversion needed."},{"id":"parasail-kimi-k2-7-code","name":"Kimi K2.7 Code","tier":"mid","context_window":null,"max_output":null,"input":0.75,"output":3.5,"cache_write":null,"cache_read":0.16,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://parasail.io/pricing","note":"Parasail elastic-endpoint serverless pricing, listed by the seller as price per 1M tokens: $0.75/1M input, $3.5/1M output, $0.16/1M cache read; no conversion needed."},{"id":"parasail-kimi-k2-6","name":"Kimi K2.6","tier":"mid","context_window":null,"max_output":null,"input":0.75,"output":3.5,"cache_write":null,"cache_read":0.16,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://parasail.io/pricing","note":"Parasail elastic-endpoint serverless pricing, listed by the seller as price per 1M tokens: $0.75/1M input, $3.5/1M output, $0.16/1M cache read; no conversion needed."},{"id":"parasail-glm-5-3-flash","name":"GLM-5.3 Flash","tier":"small","context_window":null,"max_output":null,"input":0.15,"output":0.5,"cache_write":null,"cache_read":0.03,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://parasail.io/pricing","note":"Parasail elastic-endpoint serverless pricing, listed by the seller as price per 1M tokens: $0.15/1M input, $0.5/1M output, $0.03/1M cache read; no conversion needed."},{"id":"parasail-glm-5-3","name":"GLM-5.3","tier":"flagship","context_window":null,"max_output":null,"input":1.4,"output":4.4,"cache_write":null,"cache_read":0.26,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://parasail.io/pricing","note":"Parasail elastic-endpoint serverless pricing, listed by the seller as price per 1M tokens: $1.4/1M input, $4.4/1M output, $0.26/1M cache read; no conversion needed."},{"id":"parasail-glm-5-2","name":"GLM-5.2","tier":"flagship","context_window":null,"max_output":null,"input":1.4,"output":4.4,"cache_write":null,"cache_read":0.26,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://parasail.io/pricing","note":"Parasail elastic-endpoint serverless pricing, listed by the seller as price per 1M tokens: $1.4/1M input, $4.4/1M output, $0.26/1M cache read; no conversion needed."},{"id":"parasail-glm-5-1","name":"GLM-5.1","tier":"flagship","context_window":null,"max_output":null,"input":1.4,"output":4.4,"cache_write":null,"cache_read":0.26,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://parasail.io/pricing","note":"Parasail elastic-endpoint serverless pricing, listed by the seller as price per 1M tokens: $1.4/1M input, $4.4/1M output, $0.26/1M cache read; no conversion needed."},{"id":"parasail-glm-5","name":"GLM-5","tier":"mid","context_window":null,"max_output":null,"input":1,"output":3.2,"cache_write":null,"cache_read":0.2,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://parasail.io/pricing","note":"Parasail elastic-endpoint serverless pricing, listed by the seller as price per 1M tokens: $1.0/1M input, $3.2/1M output, $0.2/1M cache read; no conversion needed."},{"id":"parasail-minimax-m3","name":"MiniMax M3","tier":"mid","context_window":null,"max_output":null,"input":0.3,"output":1.2,"cache_write":null,"cache_read":0.06,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://parasail.io/pricing","note":"Parasail elastic-endpoint serverless pricing, listed by the seller as price per 1M tokens: $0.3/1M input, $1.2/1M output, $0.06/1M cache read; no conversion needed."},{"id":"parasail-minimax-m2-5","name":"MiniMax M2.5","tier":"mid","context_window":null,"max_output":null,"input":0.3,"output":1.2,"cache_write":null,"cache_read":0.03,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://parasail.io/pricing","note":"Parasail elastic-endpoint serverless pricing, listed by the seller as price per 1M tokens: $0.3/1M input, $1.2/1M output, $0.03/1M cache read; no conversion needed."},{"id":"parasail-deepseek-v4-pro","name":"DeepSeek V4 Pro","tier":"flagship","context_window":null,"max_output":null,"input":1.74,"output":3.48,"cache_write":null,"cache_read":0.1,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://parasail.io/pricing","note":"Parasail elastic-endpoint serverless pricing, listed by the seller as price per 1M tokens: $1.74/1M input, $3.48/1M output, $0.1/1M cache read; no conversion needed."},{"id":"parasail-deepseek-v4-flash","name":"DeepSeek V4 Flash","tier":"small","context_window":null,"max_output":null,"input":0.14,"output":0.28,"cache_write":null,"cache_read":0.07,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://parasail.io/pricing","note":"Parasail elastic-endpoint serverless pricing, listed by the seller as price per 1M tokens: $0.14/1M input, $0.28/1M output, $0.07/1M cache read; no conversion needed."},{"id":"parasail-qwen3-6-35b-a3b","name":"Qwen3.6 35B-A3B","tier":"small","context_window":null,"max_output":null,"input":0.15,"output":1,"cache_write":null,"cache_read":0.05,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://parasail.io/pricing","note":"Parasail elastic-endpoint serverless pricing, listed by the seller as price per 1M tokens: $0.15/1M input, $1.0/1M output, $0.05/1M cache read; no conversion needed."},{"id":"parasail-qwen3-5-397b-a17b","name":"Qwen3.5 397B-A17B","tier":"flagship","context_window":null,"max_output":null,"input":0.5,"output":3.6,"cache_write":null,"cache_read":0.3,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://parasail.io/pricing","note":"Parasail elastic-endpoint serverless pricing, listed by the seller as price per 1M tokens: $0.5/1M input, $3.6/1M output, $0.3/1M cache read; no conversion needed."},{"id":"parasail-qwen3-5-35b-a3b","name":"Qwen3.5 35B-A3B","tier":"small","context_window":null,"max_output":null,"input":0.15,"output":1,"cache_write":null,"cache_read":0.05,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://parasail.io/pricing","note":"Parasail elastic-endpoint serverless pricing, listed by the seller as price per 1M tokens: $0.15/1M input, $1.0/1M output, $0.05/1M cache read; no conversion needed."},{"id":"parasail-qwen3-coder-next","name":"Qwen3-Coder-Next","tier":"mid","context_window":null,"max_output":null,"input":0.12,"output":0.8,"cache_write":null,"cache_read":0.07,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://parasail.io/pricing","note":"Parasail elastic-endpoint serverless pricing, listed by the seller as price per 1M tokens: $0.12/1M input, $0.8/1M output, $0.07/1M cache read; no conversion needed."},{"id":"parasail-qwen3-vl-235b-a22b","name":"Qwen3-VL 235B-A22B","tier":"flagship","context_window":null,"max_output":null,"input":0.21,"output":1.9,"cache_write":null,"cache_read":0.1,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://parasail.io/pricing","note":"Parasail elastic-endpoint serverless pricing, listed by the seller as price per 1M tokens: $0.21/1M input, $1.9/1M output, $0.1/1M cache read; no conversion needed."},{"id":"parasail-qwen3-vl-8b","name":"Qwen3-VL 8B","tier":"small","context_window":null,"max_output":null,"input":0.25,"output":0.75,"cache_write":null,"cache_read":0.12,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://parasail.io/pricing","note":"Parasail elastic-endpoint serverless pricing, listed by the seller as price per 1M tokens: $0.25/1M input, $0.75/1M output, $0.12/1M cache read; no conversion needed."},{"id":"parasail-qwen3-next-80b","name":"Qwen3-Next 80B","tier":"mid","context_window":null,"max_output":null,"input":0.1,"output":1.1,"cache_write":null,"cache_read":0.07,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://parasail.io/pricing","note":"Parasail elastic-endpoint serverless pricing, listed by the seller as price per 1M tokens: $0.1/1M input, $1.1/1M output, $0.07/1M cache read; no conversion needed."},{"id":"parasail-qwen3-235b-a22b-2507","name":"Qwen3 235B-A22B (2507)","tier":"flagship","context_window":null,"max_output":null,"input":0.14,"output":0.8,"cache_write":null,"cache_read":0.05,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://parasail.io/pricing","note":"Parasail elastic-endpoint serverless pricing, listed by the seller as price per 1M tokens: $0.14/1M input, $0.8/1M output, $0.05/1M cache read; no conversion needed."},{"id":"parasail-qwen2-5-vl-72b","name":"Qwen2.5-VL 72B","tier":"mid","context_window":null,"max_output":null,"input":0.8,"output":1,"cache_write":null,"cache_read":0.4,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://parasail.io/pricing","note":"Parasail elastic-endpoint serverless pricing, listed by the seller as price per 1M tokens: $0.8/1M input, $1.0/1M output, $0.4/1M cache read; no conversion needed."},{"id":"parasail-mistral-small-3-2-24b","name":"Mistral Small 3.2 24B","tier":"small","context_window":null,"max_output":null,"input":0.09,"output":0.3,"cache_write":null,"cache_read":0.05,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://parasail.io/pricing","note":"Parasail elastic-endpoint serverless pricing, listed by the seller as price per 1M tokens: $0.09/1M input, $0.3/1M output, $0.05/1M cache read; no conversion needed."},{"id":"parasail-llama-4-maverick-fp8","name":"Llama 4 Maverick (FP8)","tier":"mid","context_window":null,"max_output":null,"input":0.35,"output":1,"cache_write":null,"cache_read":0.17,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://parasail.io/pricing","note":"Parasail elastic-endpoint serverless pricing, listed by the seller as price per 1M tokens: $0.35/1M input, $1.0/1M output, $0.17/1M cache read; no conversion needed."},{"id":"parasail-llama-3-3-70b-fp8","name":"Llama 3.3 70B (FP8)","tier":"mid","context_window":null,"max_output":null,"input":0.22,"output":0.5,"cache_write":null,"cache_read":0.11,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://parasail.io/pricing","note":"Parasail elastic-endpoint serverless pricing, listed by the seller as price per 1M tokens: $0.22/1M input, $0.5/1M output, $0.11/1M cache read; no conversion needed."},{"id":"parasail-nemotron-3-ultra-550b-nvfp4","name":"Nemotron 3 Ultra 550B (NVFP4)","tier":"flagship","context_window":null,"max_output":null,"input":0.5,"output":2.5,"cache_write":null,"cache_read":0.1,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://parasail.io/pricing","note":"Parasail elastic-endpoint serverless pricing, listed by the seller as price per 1M tokens: $0.5/1M input, $2.5/1M output, $0.1/1M cache read; no conversion needed."},{"id":"parasail-mimo-v2-5","name":"MiMo v2.5","tier":"small","context_window":null,"max_output":null,"input":0.14,"output":0.28,"cache_write":null,"cache_read":0.05,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://parasail.io/pricing","note":"Parasail elastic-endpoint serverless pricing, listed by the seller as price per 1M tokens: $0.14/1M input, $0.28/1M output, $0.05/1M cache read; no conversion needed."},{"id":"parasail-trinity-large-thinking","name":"Trinity Large (Thinking)","tier":"reasoning","context_window":null,"max_output":null,"input":0.22,"output":0.85,"cache_write":null,"cache_read":0.06,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://parasail.io/pricing","note":"Parasail elastic-endpoint serverless pricing, listed by the seller as price per 1M tokens: $0.22/1M input, $0.85/1M output, $0.06/1M cache read; no conversion needed."},{"id":"parasail-gemma-4-26b-a4b","name":"Gemma 4 26B-A4B","tier":"small","context_window":null,"max_output":null,"input":0.13,"output":0.4,"cache_write":null,"cache_read":0.05,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://parasail.io/pricing","note":"Parasail elastic-endpoint serverless pricing, listed by the seller as price per 1M tokens: $0.13/1M input, $0.4/1M output, $0.05/1M cache read; no conversion needed."},{"id":"parasail-gemma-4-31b","name":"Gemma 4 31B","tier":"small","context_window":null,"max_output":null,"input":0.15,"output":0.4,"cache_write":null,"cache_read":0.06,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://parasail.io/pricing","note":"Parasail elastic-endpoint serverless pricing, listed by the seller as price per 1M tokens: $0.15/1M input, $0.4/1M output, $0.06/1M cache read; no conversion needed."},{"id":"parasail-skyfall-31b-v4-2","name":"Skyfall 31B v4.2","tier":"small","context_window":null,"max_output":null,"input":0.55,"output":0.8,"cache_write":null,"cache_read":0.25,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://parasail.io/pricing","note":"Parasail elastic-endpoint serverless pricing, listed by the seller as price per 1M tokens: $0.55/1M input, $0.8/1M output, $0.25/1M cache read; no conversion needed."},{"id":"parasail-cydonia-24b-v4-1","name":"Cydonia 24B v4.1","tier":"small","context_window":null,"max_output":null,"input":0.3,"output":0.5,"cache_write":null,"cache_read":0.15,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://parasail.io/pricing","note":"Parasail elastic-endpoint serverless pricing, listed by the seller as price per 1M tokens: $0.3/1M input, $0.5/1M output, $0.15/1M cache read; no conversion needed."},{"id":"parasail-gpt-oss-120b","name":"gpt-oss-120b","tier":"mid","context_window":null,"max_output":null,"input":0.1,"output":0.75,"cache_write":null,"cache_read":0.055,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://parasail.io/pricing","note":"Parasail elastic-endpoint serverless pricing, listed by the seller as price per 1M tokens: $0.1/1M input, $0.75/1M output, $0.055/1M cache read; no conversion needed."},{"id":"parasail-gpt-oss-120b-fast","name":"gpt-oss-120b (Fast)","tier":"mid","context_window":null,"max_output":null,"input":0.15,"output":0.6,"cache_write":null,"cache_read":null,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://parasail.io/pricing","note":"Parasail elastic-endpoint serverless pricing, listed by the seller as price per 1M tokens: $0.15/1M input, $0.6/1M output; no conversion needed."},{"id":"parasail-gpt-oss-20b","name":"gpt-oss-20b","tier":"small","context_window":null,"max_output":null,"input":0.04,"output":0.2,"cache_write":null,"cache_read":0.02,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://parasail.io/pricing","note":"Parasail elastic-endpoint serverless pricing, listed by the seller as price per 1M tokens: $0.04/1M input, $0.2/1M output, $0.02/1M cache read; no conversion needed."},{"id":"parasail-ui-tars-1-5-7b","name":"UI-TARS 1.5 7B","tier":"small","context_window":null,"max_output":null,"input":0.1,"output":0.2,"cache_write":null,"cache_read":0.1,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://parasail.io/pricing","note":"Parasail elastic-endpoint serverless pricing, listed by the seller as price per 1M tokens: $0.1/1M input, $0.2/1M output, $0.1/1M cache read; no conversion needed."},{"id":"parasail-gemma-3-27b","name":"Gemma 3 27B","tier":"small","context_window":null,"max_output":null,"input":0.08,"output":0.45,"cache_write":null,"cache_read":0.04,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://parasail.io/pricing","note":"Parasail elastic-endpoint serverless pricing, listed by the seller as price per 1M tokens: $0.08/1M input, $0.45/1M output, $0.04/1M cache read; no conversion needed."},{"id":"parasail-skyfall-36b-v2-fp8","name":"Skyfall 36B v2 (FP8)","tier":"small","context_window":null,"max_output":null,"input":0.55,"output":0.8,"cache_write":null,"cache_read":0.25,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://parasail.io/pricing","note":"Parasail elastic-endpoint serverless pricing, listed by the seller as price per 1M tokens: $0.55/1M input, $0.8/1M output, $0.25/1M cache read; no conversion needed."}]},{"id":"inference-net","name":"inference.net","pricing_url":"https://inference.net/models","verification":"official_page","models":[{"id":"inference-net-claude-fable-5","name":"Claude Fable 5","tier":"flagship","context_window":1000000,"max_output":128000,"input":10,"output":50,"cache_write":null,"cache_read":1,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://inference.net/models","note":"Inference.net gateway price for Claude Fable 5 (Anthropic), converted from the site's published per-token rate (costInputPerToken=0.000010, costOutputPerToken=0.000050 USD/token) to USD per 1M tokens: $10.0/1M input, $50.0/1M output, $1.0/1M cache read."},{"id":"inference-net-claude-haiku-4-5","name":"Claude Haiku 4.5","tier":"small","context_window":200000,"max_output":64000,"input":1,"output":5,"cache_write":null,"cache_read":0.1,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://inference.net/models","note":"Inference.net gateway price for Claude Haiku 4.5 (Anthropic), converted from the site's published per-token rate (costInputPerToken=0.000001, costOutputPerToken=0.000005 USD/token) to USD per 1M tokens: $1.0/1M input, $5.0/1M output, $0.1/1M cache read."},{"id":"inference-net-claude-opus-4-6","name":"Claude Opus 4.6","tier":"flagship","context_window":1000000,"max_output":128000,"input":5,"output":25,"cache_write":null,"cache_read":0.5,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://inference.net/models","note":"Inference.net gateway price for Claude Opus 4.6 (Anthropic), converted from the site's published per-token rate (costInputPerToken=0.000005, costOutputPerToken=0.000025 USD/token) to USD per 1M tokens: $5.0/1M input, $25.0/1M output, $0.5/1M cache read."},{"id":"inference-net-claude-opus-4-7","name":"Claude Opus 4.7","tier":"flagship","context_window":1000000,"max_output":128000,"input":5,"output":25,"cache_write":null,"cache_read":0.5,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://inference.net/models","note":"Inference.net gateway price for Claude Opus 4.7 (Anthropic), converted from the site's published per-token rate (costInputPerToken=0.000005, costOutputPerToken=0.000025 USD/token) to USD per 1M tokens: $5.0/1M input, $25.0/1M output, $0.5/1M cache read."},{"id":"inference-net-claude-opus-4-8","name":"Claude Opus 4.8","tier":"flagship","context_window":1000000,"max_output":128000,"input":5,"output":25,"cache_write":null,"cache_read":0.5,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://inference.net/models","note":"Inference.net gateway price for Claude Opus 4.8 (Anthropic), converted from the site's published per-token rate (costInputPerToken=0.000005, costOutputPerToken=0.000025 USD/token) to USD per 1M tokens: $5.0/1M input, $25.0/1M output, $0.5/1M cache read."},{"id":"inference-net-claude-opus-5","name":"Claude Opus 5","tier":"flagship","context_window":1000000,"max_output":128000,"input":5,"output":25,"cache_write":null,"cache_read":0.5,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://inference.net/models","note":"Inference.net gateway price for Claude Opus 5 (Anthropic), converted from the site's published per-token rate (costInputPerToken=0.000005, costOutputPerToken=0.000025 USD/token) to USD per 1M tokens: $5.0/1M input, $25.0/1M output, $0.5/1M cache read."},{"id":"inference-net-claude-sonnet-4-6","name":"Claude Sonnet 4.6","tier":"mid","context_window":1000000,"max_output":64000,"input":3,"output":15,"cache_write":null,"cache_read":0.3,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://inference.net/models","note":"Inference.net gateway price for Claude Sonnet 4.6 (Anthropic), converted from the site's published per-token rate (costInputPerToken=0.000003, costOutputPerToken=0.000015 USD/token) to USD per 1M tokens: $3.0/1M input, $15.0/1M output, $0.3/1M cache read."},{"id":"inference-net-deepseek-v4-flash","name":"DeepSeek V4 Flash","tier":"small","context_window":163840,"max_output":8192,"input":0.14,"output":0.28,"cache_write":null,"cache_read":0.0028,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://inference.net/models","note":"Inference.net gateway price for DeepSeek V4 Flash (DeepSeek), converted from the site's published per-token rate (costInputPerToken=0.000000, costOutputPerToken=0.000000 USD/token) to USD per 1M tokens: $0.14/1M input, $0.28/1M output, $0.0028/1M cache read."},{"id":"inference-net-deepseek-v4-flash-0731","name":"DeepSeek V4 Flash (0731)","tier":"small","context_window":256000,"max_output":384000,"input":0.168,"output":0.336,"cache_write":null,"cache_read":0.0336,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://inference.net/models","note":"Inference.net gateway price for DeepSeek V4 Flash (0731) (DeepSeek), converted from the site's published per-token rate (costInputPerToken=0.000000, costOutputPerToken=0.000000 USD/token) to USD per 1M tokens: $0.168/1M input, $0.336/1M output, $0.0336/1M cache read."},{"id":"inference-net-deepseek-v4-pro","name":"DeepSeek V4 Pro","tier":"flagship","context_window":1048576,"max_output":384000,"input":0.435,"output":0.87,"cache_write":null,"cache_read":0.00363,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://inference.net/models","note":"Inference.net gateway price for DeepSeek V4 Pro (DeepSeek), converted from the site's published per-token rate (costInputPerToken=0.000000, costOutputPerToken=0.000001 USD/token) to USD per 1M tokens: $0.435/1M input, $0.87/1M output, $0.00363/1M cache read."},{"id":"inference-net-deepseek-v4-pro-0813","name":"DeepSeek V4 Pro (0813)","tier":"flagship","context_window":1048576,"max_output":131072,"input":1.32,"output":3.96,"cache_write":null,"cache_read":0.044,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://inference.net/models","note":"Inference.net gateway price for DeepSeek V4 Pro (0813) (DeepSeek), converted from the site's published per-token rate (costInputPerToken=0.000001, costOutputPerToken=0.000004 USD/token) to USD per 1M tokens: $1.32/1M input, $3.96/1M output, $0.044/1M cache read."},{"id":"inference-net-gemini-2-5-flash","name":"Gemini 2.5 Flash","tier":"mid","context_window":1048576,"max_output":65536,"input":0.3,"output":2.5,"cache_write":null,"cache_read":0.03,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://inference.net/models","note":"Inference.net gateway price for Gemini 2.5 Flash (Google), converted from the site's published per-token rate (costInputPerToken=0.000000, costOutputPerToken=0.000003 USD/token) to USD per 1M tokens: $0.3/1M input, $2.5/1M output, $0.03/1M cache read."},{"id":"inference-net-gemini-2-5-flash-lite","name":"Gemini 2.5 Flash Lite","tier":"small","context_window":1048576,"max_output":65536,"input":0.1,"output":0.4,"cache_write":null,"cache_read":0.01,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://inference.net/models","note":"Inference.net gateway price for Gemini 2.5 Flash Lite (Google), converted from the site's published per-token rate (costInputPerToken=0.000000, costOutputPerToken=0.000000 USD/token) to USD per 1M tokens: $0.1/1M input, $0.4/1M output, $0.01/1M cache read."},{"id":"inference-net-gemini-2-5-pro","name":"Gemini 2.5 Pro","tier":"flagship","context_window":1048576,"max_output":65536,"input":1.25,"output":10,"cache_write":null,"cache_read":0.125,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://inference.net/models","note":"Inference.net gateway price for Gemini 2.5 Pro (Google), converted from the site's published per-token rate (costInputPerToken=0.000001, costOutputPerToken=0.000010 USD/token) to USD per 1M tokens: $1.25/1M input, $10.0/1M output, $0.125/1M cache read."},{"id":"inference-net-gemini-3-flash-preview","name":"Gemini 3 Flash Preview","tier":"mid","context_window":1048576,"max_output":65536,"input":0.5,"output":3,"cache_write":null,"cache_read":0.05,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://inference.net/models","note":"Inference.net gateway price for Gemini 3 Flash Preview (Google), converted from the site's published per-token rate (costInputPerToken=0.000000, costOutputPerToken=0.000003 USD/token) to USD per 1M tokens: $0.5/1M input, $3.0/1M output, $0.05/1M cache read."},{"id":"inference-net-gemini-3-1-flash-lite","name":"Gemini 3.1 Flash Lite","tier":"small","context_window":1048576,"max_output":65536,"input":0.25,"output":1.5,"cache_write":null,"cache_read":0.025,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://inference.net/models","note":"Inference.net gateway price for Gemini 3.1 Flash Lite (Google), converted from the site's published per-token rate (costInputPerToken=0.000000, costOutputPerToken=0.000002 USD/token) to USD per 1M tokens: $0.25/1M input, $1.5/1M output, $0.025/1M cache read."},{"id":"inference-net-gemini-3-1-pro-preview","name":"Gemini 3.1 Pro Preview","tier":"flagship","context_window":1048576,"max_output":65536,"input":2,"output":12,"cache_write":null,"cache_read":0.2,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://inference.net/models","note":"Inference.net gateway price for Gemini 3.1 Pro Preview (Google), converted from the site's published per-token rate (costInputPerToken=0.000002, costOutputPerToken=0.000012 USD/token) to USD per 1M tokens: $2.0/1M input, $12.0/1M output, $0.2/1M cache read."},{"id":"inference-net-gemini-3-5-flash","name":"Gemini 3.5 Flash","tier":"mid","context_window":1048576,"max_output":65536,"input":1.5,"output":9,"cache_write":null,"cache_read":0.15,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://inference.net/models","note":"Inference.net gateway price for Gemini 3.5 Flash (Google), converted from the site's published per-token rate (costInputPerToken=0.000002, costOutputPerToken=0.000009 USD/token) to USD per 1M tokens: $1.5/1M input, $9.0/1M output, $0.15/1M cache read."},{"id":"inference-net-gemini-3-5-flash-lite","name":"Gemini 3.5 Flash-Lite","tier":"small","context_window":1048576,"max_output":65536,"input":0.3,"output":2.5,"cache_write":null,"cache_read":0.03,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://inference.net/models","note":"Inference.net gateway price for Gemini 3.5 Flash-Lite (Google), converted from the site's published per-token rate (costInputPerToken=0.000000, costOutputPerToken=0.000003 USD/token) to USD per 1M tokens: $0.3/1M input, $2.5/1M output, $0.03/1M cache read."},{"id":"inference-net-gemini-3-6-flash","name":"Gemini 3.6 Flash","tier":"mid","context_window":1048576,"max_output":65536,"input":1.5,"output":7.5,"cache_write":null,"cache_read":0.15,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://inference.net/models","note":"Inference.net gateway price for Gemini 3.6 Flash (Google), converted from the site's published per-token rate (costInputPerToken=0.000002, costOutputPerToken=0.000008 USD/token) to USD per 1M tokens: $1.5/1M input, $7.5/1M output, $0.15/1M cache read."},{"id":"inference-net-gemini-3-7-flash","name":"Gemini 3.7 Flash","tier":"mid","context_window":1048576,"max_output":65536,"input":1.5,"output":7.5,"cache_write":null,"cache_read":0.15,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://inference.net/models","note":"Inference.net gateway price for Gemini 3.7 Flash (Google), converted from the site's published per-token rate (costInputPerToken=0.000002, costOutputPerToken=0.000008 USD/token) to USD per 1M tokens: $1.5/1M input, $7.5/1M output, $0.15/1M cache read."},{"id":"inference-net-glm-5-2","name":"GLM-5.2","tier":"flagship","context_window":1000000,"max_output":128000,"input":1.3,"output":4.4,"cache_write":null,"cache_read":0.2,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://inference.net/models","note":"Inference.net gateway price for GLM-5.2 (Z.ai), converted from the site's published per-token rate (costInputPerToken=0.000001, costOutputPerToken=0.000004 USD/token) to USD per 1M tokens: $1.3/1M input, $4.4/1M output, $0.2/1M cache read."},{"id":"inference-net-glm-5-2-fast","name":"GLM-5.2 Fast","tier":"flagship","context_window":1000000,"max_output":128000,"input":2.6,"output":8.8,"cache_write":null,"cache_read":0.4,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://inference.net/models","note":"Inference.net gateway price for GLM-5.2 Fast (Z.ai), converted from the site's published per-token rate (costInputPerToken=0.000003, costOutputPerToken=0.000009 USD/token) to USD per 1M tokens: $2.6/1M input, $8.8/1M output, $0.4/1M cache read."},{"id":"inference-net-glm-5-3","name":"GLM-5.3","tier":"flagship","context_window":1000000,"max_output":131072,"input":1.75,"output":5.5,"cache_write":null,"cache_read":0.325,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://inference.net/models","note":"Inference.net gateway price for GLM-5.3 (Z.ai), converted from the site's published per-token rate (costInputPerToken=0.000002, costOutputPerToken=0.000005 USD/token) to USD per 1M tokens: $1.75/1M input, $5.5/1M output, $0.325/1M cache read."},{"id":"inference-net-glm-5-3-flash","name":"GLM-5.3 Flash","tier":"small","context_window":1000000,"max_output":131072,"input":0.54,"output":1.8,"cache_write":null,"cache_read":0.108,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://inference.net/models","note":"Inference.net gateway price for GLM-5.3 Flash (Z.ai), converted from the site's published per-token rate (costInputPerToken=0.000001, costOutputPerToken=0.000002 USD/token) to USD per 1M tokens: $0.54/1M input, $1.8/1M output, $0.108/1M cache read."},{"id":"inference-net-gpt-4-1","name":"GPT-4.1","tier":"mid","context_window":1047576,"max_output":32768,"input":2,"output":8,"cache_write":null,"cache_read":0.5,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://inference.net/models","note":"Inference.net gateway price for GPT-4.1 (OpenAI), converted from the site's published per-token rate (costInputPerToken=0.000002, costOutputPerToken=0.000008 USD/token) to USD per 1M tokens: $2.0/1M input, $8.0/1M output, $0.5/1M cache read."},{"id":"inference-net-gpt-4-1-mini","name":"GPT-4.1 Mini","tier":"small","context_window":1047576,"max_output":32768,"input":0.4,"output":1.6,"cache_write":null,"cache_read":0.1,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://inference.net/models","note":"Inference.net gateway price for GPT-4.1 Mini (OpenAI), converted from the site's published per-token rate (costInputPerToken=0.000000, costOutputPerToken=0.000002 USD/token) to USD per 1M tokens: $0.4/1M input, $1.6/1M output, $0.1/1M cache read."},{"id":"inference-net-gpt-4-1-nano","name":"GPT-4.1 Nano","tier":"small","context_window":1047576,"max_output":32768,"input":0.1,"output":0.4,"cache_write":null,"cache_read":0.025,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://inference.net/models","note":"Inference.net gateway price for GPT-4.1 Nano (OpenAI), converted from the site's published per-token rate (costInputPerToken=0.000000, costOutputPerToken=0.000000 USD/token) to USD per 1M tokens: $0.1/1M input, $0.4/1M output, $0.025/1M cache read."},{"id":"inference-net-gpt-4o","name":"GPT-4o","tier":"mid","context_window":128000,"max_output":16384,"input":2.5,"output":10,"cache_write":null,"cache_read":1.25,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://inference.net/models","note":"Inference.net gateway price for GPT-4o (OpenAI), converted from the site's published per-token rate (costInputPerToken=0.000003, costOutputPerToken=0.000010 USD/token) to USD per 1M tokens: $2.5/1M input, $10.0/1M output, $1.25/1M cache read."},{"id":"inference-net-gpt-4o-mini","name":"GPT-4o Mini","tier":"small","context_window":128000,"max_output":16384,"input":0.15,"output":0.6,"cache_write":null,"cache_read":0.075,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://inference.net/models","note":"Inference.net gateway price for GPT-4o Mini (OpenAI), converted from the site's published per-token rate (costInputPerToken=0.000000, costOutputPerToken=0.000001 USD/token) to USD per 1M tokens: $0.15/1M input, $0.6/1M output, $0.075/1M cache read."},{"id":"inference-net-gpt-5-mini","name":"GPT-5 Mini","tier":"small","context_window":400000,"max_output":128000,"input":0.25,"output":2,"cache_write":null,"cache_read":0.025,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://inference.net/models","note":"Inference.net gateway price for GPT-5 Mini (OpenAI), converted from the site's published per-token rate (costInputPerToken=0.000000, costOutputPerToken=0.000002 USD/token) to USD per 1M tokens: $0.25/1M input, $2.0/1M output, $0.025/1M cache read."},{"id":"inference-net-gpt-5-nano","name":"GPT-5 Nano","tier":"small","context_window":400000,"max_output":128000,"input":0.05,"output":0.4,"cache_write":null,"cache_read":0.005,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://inference.net/models","note":"Inference.net gateway price for GPT-5 Nano (OpenAI), converted from the site's published per-token rate (costInputPerToken=0.000000, costOutputPerToken=0.000000 USD/token) to USD per 1M tokens: $0.05/1M input, $0.4/1M output, $0.005/1M cache read."},{"id":"inference-net-gpt-5-4","name":"GPT-5.4","tier":"flagship","context_window":1050000,"max_output":128000,"input":2.5,"output":15,"cache_write":null,"cache_read":0.25,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://inference.net/models","note":"Inference.net gateway price for GPT-5.4 (OpenAI), converted from the site's published per-token rate (costInputPerToken=0.000003, costOutputPerToken=0.000015 USD/token) to USD per 1M tokens: $2.5/1M input, $15.0/1M output, $0.25/1M cache read."},{"id":"inference-net-gpt-5-5","name":"GPT-5.5","tier":"flagship","context_window":1050000,"max_output":128000,"input":5,"output":30,"cache_write":null,"cache_read":0.5,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://inference.net/models","note":"Inference.net gateway price for GPT-5.5 (OpenAI), converted from the site's published per-token rate (costInputPerToken=0.000005, costOutputPerToken=0.000030 USD/token) to USD per 1M tokens: $5.0/1M input, $30.0/1M output, $0.5/1M cache read."},{"id":"inference-net-gpt-5-6-luna","name":"GPT-5.6 Luna","tier":"small","context_window":1050000,"max_output":128000,"input":0.2,"output":1.2,"cache_write":null,"cache_read":0.02,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://inference.net/models","note":"Inference.net gateway price for GPT-5.6 Luna (OpenAI), converted from the site's published per-token rate (costInputPerToken=0.000000, costOutputPerToken=0.000001 USD/token) to USD per 1M tokens: $0.2/1M input, $1.2/1M output, $0.02/1M cache read."},{"id":"inference-net-gpt-5-6-sol","name":"GPT-5.6 Sol","tier":"flagship","context_window":1050000,"max_output":128000,"input":5,"output":30,"cache_write":null,"cache_read":0.5,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://inference.net/models","note":"Inference.net gateway price for GPT-5.6 Sol (OpenAI), converted from the site's published per-token rate (costInputPerToken=0.000005, costOutputPerToken=0.000030 USD/token) to USD per 1M tokens: $5.0/1M input, $30.0/1M output, $0.5/1M cache read."},{"id":"inference-net-gpt-5-6-terra","name":"GPT-5.6 Terra","tier":"mid","context_window":1050000,"max_output":128000,"input":2,"output":12,"cache_write":null,"cache_read":0.2,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://inference.net/models","note":"Inference.net gateway price for GPT-5.6 Terra (OpenAI), converted from the site's published per-token rate (costInputPerToken=0.000002, costOutputPerToken=0.000012 USD/token) to USD per 1M tokens: $2.0/1M input, $12.0/1M output, $0.2/1M cache read."},{"id":"inference-net-grok-4-5","name":"Grok 4.5","tier":"mid","context_window":500000,"max_output":500000,"input":2,"output":6,"cache_write":null,"cache_read":0.3,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://inference.net/models","note":"Inference.net gateway price for Grok 4.5 (xAI), converted from the site's published per-token rate (costInputPerToken=0.000002, costOutputPerToken=0.000006 USD/token) to USD per 1M tokens: $2.0/1M input, $6.0/1M output, $0.3/1M cache read."},{"id":"inference-net-grok-4-6","name":"Grok 4.6","tier":"mid","context_window":500000,"max_output":500000,"input":2,"output":6,"cache_write":null,"cache_read":0.5,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://inference.net/models","note":"Inference.net gateway price for Grok 4.6 (xAI), converted from the site's published per-token rate (costInputPerToken=0.000002, costOutputPerToken=0.000006 USD/token) to USD per 1M tokens: $2.0/1M input, $6.0/1M output, $0.5/1M cache read."},{"id":"inference-net-kimi-k3","name":"Kimi K3","tier":"flagship","context_window":1048576,"max_output":1048576,"input":3,"output":15,"cache_write":null,"cache_read":0.3,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://inference.net/models","note":"Inference.net gateway price for Kimi K3 (Moonshot AI), converted from the site's published per-token rate (costInputPerToken=0.000003, costOutputPerToken=0.000015 USD/token) to USD per 1M tokens: $3.0/1M input, $15.0/1M output, $0.3/1M cache read."},{"id":"inference-net-kimi-k3-fast","name":"Kimi K3 Fast","tier":"flagship","context_window":1048576,"max_output":1048576,"input":4.5,"output":22.5,"cache_write":null,"cache_read":0.45,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://inference.net/models","note":"Inference.net gateway price for Kimi K3 Fast (Moonshot AI), converted from the site's published per-token rate (costInputPerToken=0.000005, costOutputPerToken=0.000023 USD/token) to USD per 1M tokens: $4.5/1M input, $22.5/1M output, $0.45/1M cache read."},{"id":"inference-net-schematron-v2-small","name":"schematron-v2-small","tier":"small","context_window":128000,"max_output":4096,"input":0.05,"output":0.25,"cache_write":null,"cache_read":0.05,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://inference.net/models","note":"Inference.net gateway price for schematron-v2-small (InferenceNet), converted from the site's published per-token rate (costInputPerToken=0.000000, costOutputPerToken=0.000000 USD/token) to USD per 1M tokens: $0.05/1M input, $0.25/1M output, $0.05/1M cache read."},{"id":"inference-net-schematron-v2-turbo","name":"schematron-v2-turbo","tier":"small","context_window":128000,"max_output":8192,"input":0.03,"output":0.15,"cache_write":null,"cache_read":0.03,"modalities":["text"],"released":null,"verified":"2026-08-30","source":"https://inference.net/models","note":"Inference.net gateway price for schematron-v2-turbo (InferenceNet), converted from the site's published per-token rate (costInputPerToken=0.000000, costOutputPerToken=0.000000 USD/token) to USD per 1M tokens: $0.03/1M input, $0.15/1M output, $0.03/1M cache read."}]}],"verification_levels":{"official_page":"Fetched directly from the provider's official pricing page and read off it.","cross_referenced":"Cross-referenced across multiple independent sources that quote the official page, but the official page itself was not reachable from our build environment. Treat as high-confidence but unconfirmed at the primary source.","official_api":"Retrieved directly from the provider's own public pricing API as structured, machine-readable data rather than read off a rendered page. High-confidence on the numbers themselves; note that where this applies to OpenRouter, OpenRouter is a reseller, so its price is what you pay through OpenRouter specifically, which can differ from the originating lab's own price."},"families":{"deepseek-v4-pro":"DeepSeek V4 Pro","glm-5.3":"GLM-5.3","gpt-oss-120b":"gpt-oss-120b","gpt-oss-20b":"gpt-oss-20b","kimi-k3":"Kimi K3","llama-3.1-8b":"Llama 3.1 8B","llama-3.3-70b":"Llama 3.3 70B","llama-4-maverick":"Llama 4 Maverick","qwen3.8-max":"Qwen3.8 Max"},"license":"CC BY 4.0","license_url":"https://creativecommons.org/licenses/by/4.0/","attribution":"https://is.kluuz.de/"}