Replicate API pricing

30 models from Replicate, verified 2026-08-30. Official pricing page →

Official page. Fetched directly from the provider's official pricing page and read off it.

Models and prices

ModelTierContext$/1M in$/1M out$/1M cache read
anthropic/claude-opus-4.6Flagship—$5.00$25.00—
google/gemini-3-proFlagship—$2.00$12.00—
google/gemini-3.1-proFlagship—$2.00$12.00—
openai/gpt-5Flagship—$1.25$10.00—
openai/gpt-5.2Flagship—$1.75$14.00—
deepseek-ai/deepseek-r1Reasoning—$3.75$10.00—
openai/o1Reasoning—$15.00$60.00—
openai/o1-miniReasoning—$1.10$4.40—
openai/o4-miniReasoning—$1.00$4.00—
anthropic/claude-4-sonnetMid—$3.00$15.00—
anthropic/claude-4.5-sonnetMid—$3.00$15.00—
deepseek-ai/deepseek-v3Mid—$1.45$1.45—
deepseek-ai/deepseek-v3.1Mid—$0.672$2.02—
meta/meta-llama-3-70b-instructMid—$0.65$2.75—
moonshotai/kimi-k2.5Mid—$0.72$3.60—
openai/gpt-4.1Mid—$2.00$8.00—
openai/gpt-4oMid—$2.50$10.00—
openai/gpt-5-miniMid—$0.25$2.00—
openai/gpt-oss-120bMid—$0.18$0.72—
qwen/qwen3-235b-a22b-instruct-2507Mid—$0.264$1.06—
anthropic/claude-3.5-haikuSmall—$1.00$5.00—
anthropic/claude-4.5-haikuSmall—$1.00$5.00—
google/gemini-2.5-flashSmall—$0.30$2.50—
ibm-granite/granite-3.3-8b-instructSmall—$0.03$0.25—
meta/meta-llama-3-8b-instructSmall—$0.05$0.25—
openai/gpt-4.1-miniSmall—$0.40$1.60—
openai/gpt-4.1-nanoSmall—$0.10$0.40—
openai/gpt-4o-miniSmall—$0.15$0.60—
openai/gpt-5-nanoSmall—$0.05$0.40—
openai/gpt-oss-20bSmall—$0.09$0.36—