LLM API Cost Calculator
All models Compare Where to run Price changes Price history Price log API

DeepInfra API pricing history

Every DeepInfra list price we can evidence, back to August 24, 2024 — 25 dated observations across 11 models. Each row links to the page the figure was read off, usually a timestamped snapshot in the Internet Archive. Plus 18 more points recorded by LiteLLM, an aggregator, across 8 of the models above — not seller-verified, marked as such in the tables below.

Steepest evidenced fall here: Llama-3.1-405B-Instruct, $2.70 to $0.80 per 1M output tokens — down 70% in 192 days.

Price per 1M output tokens over time, logarithmic scale Mixtral-8x7B-Instruct-v0.1: $0.24 on August 24, 2024 to $0.24 on August 21, 2025; Llama-3.1-8B-Instruct: $0.055 on August 24, 2024 to $0.05 on August 21, 2025; Llama-3.1-70B-Instruct: $0.40 on August 24, 2024 to $0.40 on August 21, 2025; Llama-3.1-405B-Instruct: $2.70 on August 24, 2024 to $0.80 on March 4, 2025; Llama-4-Scout-17B-16E: $0.30 on August 21, 2025 to $0.30 on August 2, 2026; Llama-4-Maverick-17B-128E: $0.60 on August 21, 2025 to $0.80 on August 2, 2026; Llama-3.3-70B-Instruct-Turbo: $0.12 on August 21, 2025 to $0.32 on August 2, 2026; Meta-Llama-3.1-8B-Instruct-Turbo: $0.04 on August 2, 2026, a single observation $0.01$0.03$0.10$0.30$1.00$3.00$10.0020252026 Mixtral-8x7B-Instruct-v0.1 — August 24, 2024 — $0.24 per 1M output tokensMixtral-8x7B-Instruct-v0.1 — December 3, 2024 — $0.24 per 1M output tokensMixtral-8x7B-Instruct-v0.1 — March 4, 2025 — $0.24 per 1M output tokensMixtral-8x7B-Instruct-v0.1 — August 21, 2025 — $0.24 per 1M output tokensLlama-3.1-8B-Instruct — August 24, 2024 — $0.055 per 1M output tokensLlama-3.1-8B-Instruct — December 3, 2024 — $0.055 per 1M output tokensLlama-3.1-8B-Instruct — March 4, 2025 — $0.05 per 1M output tokensLlama-3.1-8B-Instruct — August 21, 2025 — $0.05 per 1M output tokensLlama-3.1-70B-Instruct — August 24, 2024 — $0.40 per 1M output tokensLlama-3.1-70B-Instruct — December 3, 2024 — $0.40 per 1M output tokensLlama-3.1-70B-Instruct — March 4, 2025 — $0.40 per 1M output tokensLlama-3.1-70B-Instruct — August 21, 2025 — $0.40 per 1M output tokensLlama-3.1-405B-Instruct — August 24, 2024 — $2.70 per 1M output tokensLlama-3.1-405B-Instruct — December 3, 2024 — $1.79 per 1M output tokensLlama-3.1-405B-Instruct — March 4, 2025 — $0.80 per 1M output tokensLlama-4-Scout-17B-16E — August 21, 2025 — $0.30 per 1M output tokensLlama-4-Scout-17B-16E — August 2, 2026 — $0.30 per 1M output tokensLlama-4-Maverick-17B-128E — August 21, 2025 — $0.60 per 1M output tokensLlama-4-Maverick-17B-128E — August 2, 2026 — $0.80 per 1M output tokensLlama-3.3-70B-Instruct-Turbo — August 21, 2025 — $0.12 per 1M output tokensLlama-3.3-70B-Instruct-Turbo — August 2, 2026 — $0.32 per 1M output tokensMeta-Llama-3.1-8B-Instruct-Turbo — August 2, 2026 — $0.04 per 1M output tokens
  • Mixtral-8x7B-Instruct-v0.1
  • Llama-3.1-8B-Instruct
  • Llama-3.1-70B-Instruct
  • Llama-3.1-405B-Instruct
  • Llama-4-Scout-17B-16E
  • Llama-4-Maverick-17B-128E
  • Llama-3.3-70B-Instruct-Turbo
  • Meta-Llama-3.1-8B-Instruct-Turbo (one observation)
Showing the 8 most-observed models; 3 more are in the dated tables below.

How to read this, and what it is not

  • These are list prices, standard synchronous tier. Batch, cached-input and off-peak discounts are excluded, so real bills are often lower than the line.
  • A date is an observation, not a change date. A snapshot dated March 2024 means the page said that in March 2024 — the price may have taken effect earlier. Rows marked Launch or Price change do carry the vendor's own date.
  • Rows marked "LiteLLM, monthly" are not seller-verified. They are read from a dated commit of LiteLLM's aggregated price file, not from the provider's own page — a weaker guarantee than every other row here. Their date is the month a price was first recorded in that monthly snapshot, never the day it changed, and the true change could fall anywhere in the weeks before it. Each one still links to the exact commit it came from, so you can check it yourself.
  • Gaps are gaps. Lines connect observations; they are not evidence about the days in between, and a missing model means nobody has evidenced it yet, not that its price held.
  • Same name, different thing. Models are quietly revised under a stable name. A flat line can hide a model that got better at the same price — which is itself a price cut per unit of capability, and one this dataset cannot see.
  • Prices in CNY are shown in CNY. Converting an old observation with today's exchange rate would invent a number, so we don't.

Llama-3-70B-Instruct

DateInput / 1MOutput / 1MEvidenceLink
$0.35$0.40 Pricing page source

Llama-3-8B-Instruct

DateInput / 1MOutput / 1MEvidenceLink
$0.055$0.055 Pricing page (blended) source

Llama-3.1-405B-Instruct

DateInput / 1MOutput / 1MEvidenceLink
$2.70$2.70 Launch (blended) source
$1.79$1.79 Price change (blended) source
$0.80$0.80 Price change (blended) source

Output price fell 70% over 192 days, from $2.70 to $0.80 per 1M tokens.

Llama-3.1-70B-Instruct

DateInput / 1MOutput / 1MEvidenceLink
$0.35$0.40 Pricing page source
$0.35$0.40 Pricing page source
$0.23$0.40 Price change source
$0.23$0.40 Pricing page source
$0.23$0.40 First recorded — LiteLLM, monthly source
$0.40$0.40 Price changed — LiteLLM, monthly source

Output price unchanged across 362 days of observations.

2 rows marked LiteLLM, monthly above are recorded by an aggregator, not read from DeepInfra's own page — see "How to read this" above.

Llama-3.1-8B-Instruct

DateInput / 1MOutput / 1MEvidenceLink
$0.055$0.055 Pricing page (blended) source
$0.055$0.055 Pricing page (blended) source
$0.03$0.05 Price change source
$0.03$0.05 Pricing page source
$0.03$0.05 First recorded — LiteLLM, monthly source

Output price fell 9% over 362 days, from $0.055 to $0.05 per 1M tokens.

1 row marked LiteLLM, monthly above is recorded by an aggregator, not read from DeepInfra's own page — see "How to read this" above.

Llama-3.3-70B-Instruct-Turbo

DateInput / 1MOutput / 1MEvidenceLink
$0.038$0.12 Pricing page source
$0.038$0.12 First recorded — LiteLLM, monthly source
$0.13$0.39 Price changed — LiteLLM, monthly source
$0.10$0.32 Price change source
$0.10$0.32 Price changed — LiteLLM, monthly source

Output price rose 167% over 346 days, from $0.12 to $0.32 per 1M tokens.

3 rows marked LiteLLM, monthly above are recorded by an aggregator, not read from DeepInfra's own page — see "How to read this" above.

Llama-4-Maverick-17B-128E

DateInput / 1MOutput / 1MEvidenceLink
$0.15$0.60 Launch source
$0.15$0.60 First recorded — LiteLLM, monthly source
$0.20$0.80 Price change source
$0.20$0.80 Price changed — LiteLLM, monthly source

Output price rose 33% over 346 days, from $0.60 to $0.80 per 1M tokens.

2 rows marked LiteLLM, monthly above are recorded by an aggregator, not read from DeepInfra's own page — see "How to read this" above.

Llama-4-Scout-17B-16E

DateInput / 1MOutput / 1MEvidenceLink
$0.08$0.30 Launch source
$0.08$0.30 First recorded — LiteLLM, monthly source
$0.10$0.30 Price change source
$0.10$0.30 Price changed — LiteLLM, monthly source

Output price unchanged across 346 days of observations.

2 rows marked LiteLLM, monthly above are recorded by an aggregator, not read from DeepInfra's own page — see "How to read this" above.

Meta-Llama-3.1-70B-Instruct-Turbo

DateInput / 1MOutput / 1MEvidenceLink
$0.10$0.28 First recorded — LiteLLM, monthly source
$0.40$0.40 Pricing page (blended) source
$0.40$0.40 Price changed — LiteLLM, monthly source

2 rows marked LiteLLM, monthly above are recorded by an aggregator, not read from DeepInfra's own page — see "How to read this" above.

Meta-Llama-3.1-8B-Instruct-Turbo

DateInput / 1MOutput / 1MEvidenceLink
$0.015$0.02 First recorded — LiteLLM, monthly source
$0.02$0.03 Price changed — LiteLLM, monthly source
$0.02$0.04 Pricing page source
$0.02$0.04 Price changed — LiteLLM, monthly source

3 rows marked LiteLLM, monthly above are recorded by an aggregator, not read from DeepInfra's own page — see "How to read this" above.

Mixtral-8x7B-Instruct-v0.1

DateInput / 1MOutput / 1MEvidenceLink
$0.27$0.27 First recorded — LiteLLM, monthly source
$0.24$0.24 Pricing page (blended) source
$0.24$0.24 Pricing page (blended) source
$0.24$0.24 Pricing page (blended) source
$0.08$0.24 Price change source
$0.08$0.24 Price changed — LiteLLM, monthly source
$0.40$0.40 Price changed — LiteLLM, monthly source

Output price unchanged across 362 days of observations.

3 rows marked LiteLLM, monthly above are recorded by an aggregator, not read from DeepInfra's own page — see "How to read this" above.

Back to every provider's price history · take the data: price-history.json (CC BY 4.0).