Free tool

LLM API pricing, per million tokens

Input, cached input, output and batch prices for OpenAI, Anthropic, Google and Mistral models, in US dollars per 1 million tokens. Every row links to the vendor's own pricing page and says when we checked it.

Prices checked 2026-11-12. Confirm on OpenAI's pricing page, Anthropic's pricing page, Google's pricing page and Mistral's pricing page.

Provider

Showing 63 of 63 models. Click a column heading to sort.

LLM API prices in US dollars per 1 million tokens
Source
gpt-6-astra
Notes
  • Cache write: $12.50.
  • The pricing page has a Short (272K input tokens or fewer) / Long (over 272K) switch. Only the Short prices are in the page source, so longer prompts are not priced here.
  • Cache write is OpenAI's 30-minute write price.
  • Context window from the model page.
OpenAI$10.00$1.00$50.00$5.00$25.001.05M$0.15Pricing pagechecked 2026-11-12
gpt-6.1-sol
Notes
  • Cache write: $2.50.
  • The pricing page has a Short (272K input tokens or fewer) / Long (over 272K) switch. Only the Short prices are in the page source, so longer prompts are not priced here.
  • Cache write is OpenAI's 30-minute write price.
  • Context window from the model page.
OpenAI$2.00$0.10$10.00$1.00$5.001.05M$0.03Pricing pagechecked 2026-11-12
gpt-6.1-terra
Notes
  • Cache write: $1.25.
  • The pricing page has a Short (272K input tokens or fewer) / Long (over 272K) switch. Only the Short prices are in the page source, so longer prompts are not priced here.
  • Cache write is OpenAI's 30-minute write price.
  • Context window from the model page.
OpenAI$1.00$0.05$5.00$0.50$2.501.05M$0.015Pricing pagechecked 2026-11-12
gpt-6-luna
Notes
  • Cache write: $0.125.
  • The pricing page has a Short (272K input tokens or fewer) / Long (over 272K) switch. Only the Short prices are in the page source, so longer prompts are not priced here.
  • Cache write is OpenAI's 30-minute write price.
  • Context window from the model page.
OpenAI$0.10$0.01$0.50$0.05$0.251.05M$0.0015Pricing pagechecked 2026-11-12
gpt-6-sol
Notes
  • Cache write: $2.50.
  • The pricing page has a Short (272K input tokens or fewer) / Long (over 272K) switch. Only the Short prices are in the page source, so longer prompts are not priced here.
  • Cache write is OpenAI's 30-minute write price.
  • Context window from the model page.
OpenAI$2.00$0.20$10.00$1.00$5.001.05M$0.03Pricing pagechecked 2026-11-12
gpt-5.6-sol
Notes
  • Listed price ends after 2026-11-30.
  • Cache write: $5.00.
  • The pricing page has a Short (272K input tokens or fewer) / Long (over 272K) switch. Only the Short prices are in the page source, so longer prompts are not priced here.
  • Cache write is OpenAI's 30-minute write price.
  • OpenAI says this is promotional pricing, available until November 30, 2026. The price after that is not on the page.
  • Context window from the model page.
OpenAI$4.00$0.40$20.00$2.00$10.001.05M$0.06Pricing pagechecked 2026-11-12
gpt-5.6-terra
Notes
  • Cache write: $2.50.
  • The pricing page has a Short (272K input tokens or fewer) / Long (over 272K) switch. Only the Short prices are in the page source, so longer prompts are not priced here.
  • Cache write is OpenAI's 30-minute write price.
  • Context window from the model page.
OpenAI$2.00$0.20$12.00$1.00$6.001.05M$0.032Pricing pagechecked 2026-11-12
gpt-5.6-luna
Notes
  • Cache write: $0.25.
  • The pricing page has a Short (272K input tokens or fewer) / Long (over 272K) switch. Only the Short prices are in the page source, so longer prompts are not priced here.
  • Cache write is OpenAI's 30-minute write price.
  • Context window from the model page.
OpenAI$0.20$0.02$1.20$0.10$0.601.05M$0.0032Pricing pagechecked 2026-11-12
gpt-5.5
Notes
  • OpenAI lists this price for prompts under 272K tokens.
  • The pricing page has a Short (272K input tokens or fewer) / Long (over 272K) switch. Only the Short prices are in the page source, so longer prompts are not priced here.
  • Context window from the model page.
OpenAI$5.00$0.50$30.00$2.50$15.001.05M$0.08Pricing pagechecked 2026-11-12
gpt-5.5-pro
Notes
  • OpenAI lists this price for prompts under 272K tokens.
  • The pricing page has a Short (272K input tokens or fewer) / Long (over 272K) switch. Only the Short prices are in the page source, so longer prompts are not priced here.
  • Context window from the model page.
OpenAI$30.00n/a$180.00$15.00$90.001.05M$0.48Pricing pagechecked 2026-11-12
gpt-5.4
Notes
  • OpenAI lists this price for prompts under 272K tokens.
  • The pricing page has a Short (272K input tokens or fewer) / Long (over 272K) switch. Only the Short prices are in the page source, so longer prompts are not priced here.
  • Context window from the model page.
OpenAI$2.50$0.25$15.00$1.25$7.501.05M$0.04Pricing pagechecked 2026-11-12
gpt-5.4-mini
Notes
  • The pricing page has a Short (272K input tokens or fewer) / Long (over 272K) switch. Only the Short prices are in the page source, so longer prompts are not priced here.
  • Context window from the model page.
OpenAI$0.75$0.075$4.50$0.375$2.25400K$0.012Pricing pagechecked 2026-11-12
gpt-5.4-nano
Notes
  • The pricing page has a Short (272K input tokens or fewer) / Long (over 272K) switch. Only the Short prices are in the page source, so longer prompts are not priced here.
  • Context window from the model page.
OpenAI$0.20$0.02$1.25$0.10$0.625400K$0.00325Pricing pagechecked 2026-11-12
gpt-5.4-pro
Notes
  • OpenAI lists this price for prompts under 272K tokens.
  • The pricing page has a Short (272K input tokens or fewer) / Long (over 272K) switch. Only the Short prices are in the page source, so longer prompts are not priced here.
  • Context window from the model page.
OpenAI$30.00n/a$180.00$15.00$90.001.05M$0.48Pricing pagechecked 2026-11-12
gpt-5.2
Notes
  • The pricing page has a Short (272K input tokens or fewer) / Long (over 272K) switch. Only the Short prices are in the page source, so longer prompts are not priced here.
  • Context window from the model page.
OpenAI$1.75$0.175$14.00$0.875$7.00400K$0.0315Pricing pagechecked 2026-11-12
gpt-5.2-pro
Notes
  • The pricing page has a Short (272K input tokens or fewer) / Long (over 272K) switch. Only the Short prices are in the page source, so longer prompts are not priced here.
  • Context window from the model page.
OpenAI$21.00n/a$168.00$10.50$84.00400K$0.378Pricing pagechecked 2026-11-12
gpt-5.1
Notes
  • The pricing page has a Short (272K input tokens or fewer) / Long (over 272K) switch. Only the Short prices are in the page source, so longer prompts are not priced here.
  • Context window from the model page.
OpenAI$1.25$0.125$10.00$0.625$5.00400K$0.0225Pricing pagechecked 2026-11-12
gpt-5
Notes
  • The pricing page has a Short (272K input tokens or fewer) / Long (over 272K) switch. Only the Short prices are in the page source, so longer prompts are not priced here.
  • Context window from the model page.
OpenAI$1.25$0.125$10.00$0.625$5.00400K$0.0225Pricing pagechecked 2026-11-12
gpt-5-mini
Notes
  • The pricing page has a Short (272K input tokens or fewer) / Long (over 272K) switch. Only the Short prices are in the page source, so longer prompts are not priced here.
  • Context window from the model page.
OpenAI$0.25$0.025$2.00$0.125$1.00400K$0.0045Pricing pagechecked 2026-11-12
gpt-5-nano
Notes
  • The pricing page has a Short (272K input tokens or fewer) / Long (over 272K) switch. Only the Short prices are in the page source, so longer prompts are not priced here.
  • Context window from the model page.
OpenAI$0.05$0.005$0.40$0.025$0.20400K$0.0009Pricing pagechecked 2026-11-12
gpt-5-pro
Notes
  • The pricing page has a Short (272K input tokens or fewer) / Long (over 272K) switch. Only the Short prices are in the page source, so longer prompts are not priced here.
  • Context window from the model page.
OpenAI$15.00n/a$120.00$7.50$60.00400K$0.27Pricing pagechecked 2026-11-12
gpt-4.1
Notes
  • The pricing page has a Short (272K input tokens or fewer) / Long (over 272K) switch. Only the Short prices are in the page source, so longer prompts are not priced here.
  • Context window from the model page.
OpenAI$2.00$0.50$8.00$1.00$4.001.05M$0.028Pricing pagechecked 2026-11-12
gpt-4.1-mini
Notes
  • The pricing page has a Short (272K input tokens or fewer) / Long (over 272K) switch. Only the Short prices are in the page source, so longer prompts are not priced here.
  • Context window from the model page.
OpenAI$0.40$0.10$1.60$0.20$0.801.05M$0.0056Pricing pagechecked 2026-11-12
gpt-4o
Notes
OpenAI$2.50$1.25$10.00$1.25$5.00128K$0.035Pricing pagechecked 2026-11-12
gpt-4o-mini
Notes
OpenAI$0.15$0.075$0.60$0.075$0.30128K$0.0021Pricing pagechecked 2026-11-12
o3-pro
Notes
OpenAI$20.00n/a$80.00$10.00$40.00200K$0.28Pricing pagechecked 2026-11-12
o3
Notes
OpenAI$2.00$0.50$8.00$1.00$4.00200K$0.028Pricing pagechecked 2026-11-12
Claude Fable 5.2claude-fable-5-2
Notes
  • Cache write: $10.00.
  • Cache write is Anthropic's 5-minute write price; a 1-hour write costs $16.00.
  • Cached input is Anthropic's cache hits and refreshes price.
  • Context window from the model page.
Anthropic$8.00$0.20$40.00$4.00$20.001M$0.12Pricing pagechecked 2026-11-12
Claude Opus 5.5claude-opus-5-5
Notes
  • Cache write: $5.00.
  • Cache write is Anthropic's 5-minute write price; a 1-hour write costs $8.00.
  • Cached input is Anthropic's cache hits and refreshes price.
  • Context window from the model page.
Anthropic$4.00$0.20$20.00$2.00$10.001M$0.06Pricing pagechecked 2026-11-12
Claude Sonnet 5.5claude-sonnet-5-5
Notes
  • Cache write: $2.50.
  • Cache write is Anthropic's 5-minute write price; a 1-hour write costs $4.00.
  • Cached input is Anthropic's cache hits and refreshes price.
  • Context window from the model page.
Anthropic$2.00$0.10$10.00$1.00$5.001M$0.03Pricing pagechecked 2026-11-12
Claude Haiku 5.5claude-haiku-5-5
Notes
  • Prompts over 100,000 tokens: input $0.50, cached input $0.05, cache write $0.625, output $2.50, batch input $0.25, batch output $1.25.
  • Cache write: $0.125.
  • Cache write is Anthropic's 5-minute write price; a 1-hour write costs $0.20.
  • Cached input is Anthropic's cache hits and refreshes price.
  • Prompts over 100,000 tokens (cache reads and writes included) pay the higher prices.
  • Context window from the model page.
Anthropic$0.10$0.01$0.50$0.05$0.251M$0.0015Pricing pagechecked 2026-11-12
Claude Mythos 5.2
Notes
  • Cache write: $10.00.
  • Cache write is Anthropic's 5-minute write price; a 1-hour write costs $16.00.
  • Cached input is Anthropic's cache hits and refreshes price.
  • Anthropic's models overview marks this model: verification required.
  • Context window from the model page.
Anthropic$8.00$0.20$40.00$4.00$20.001M$0.12Pricing pagechecked 2026-11-12
Claude Fable 5.1
Notes
  • Cache write: $12.50.
  • Cache write is Anthropic's 5-minute write price; a 1-hour write costs $20.00.
  • Cached input is Anthropic's cache hits and refreshes price.
  • Context window from the model page.
Anthropic$10.00$0.25$50.00$5.00$25.001M$0.15Pricing pagechecked 2026-11-12
Claude Mythos 5.1
Notes
  • Cache write: $12.50.
  • Cache write is Anthropic's 5-minute write price; a 1-hour write costs $20.00.
  • Cached input is Anthropic's cache hits and refreshes price.
  • Anthropic's models overview marks this model: verification required.
  • Context window from the model page.
Anthropic$10.00$0.25$50.00$5.00$25.001M$0.15Pricing pagechecked 2026-11-12
Claude Fable 5
Notes
  • Cache write: $12.50.
  • Cache write is Anthropic's 5-minute write price; a 1-hour write costs $20.00.
  • Cached input is Anthropic's cache hits and refreshes price.
  • Context window from the model page.
Anthropic$10.00$1.00$50.00$5.00$25.001M$0.15Pricing pagechecked 2026-11-12
Claude Mythos 5
Notes
  • Cache write: $12.50.
  • Cache write is Anthropic's 5-minute write price; a 1-hour write costs $20.00.
  • Cached input is Anthropic's cache hits and refreshes price.
  • Anthropic's models overview marks this model: verification required.
  • Context window from the model page.
Anthropic$10.00$1.00$50.00$5.00$25.001M$0.15Pricing pagechecked 2026-11-12
Claude Opus 5
Notes
  • Cache write: $6.25.
  • Cache write is Anthropic's 5-minute write price; a 1-hour write costs $10.00.
  • Cached input is Anthropic's cache hits and refreshes price.
  • Context window from the model page.
Anthropic$5.00$0.50$25.00$2.50$12.501M$0.075Pricing pagechecked 2026-11-12
Claude Opus 4.8
Notes
  • Cache write: $6.25.
  • Cache write is Anthropic's 5-minute write price; a 1-hour write costs $10.00.
  • Cached input is Anthropic's cache hits and refreshes price.
  • Context window from the model page.
Anthropic$5.00$0.50$25.00$2.50$12.501M$0.075Pricing pagechecked 2026-11-12
Claude Opus 4.7
Notes
  • Cache write: $6.25.
  • Cache write is Anthropic's 5-minute write price; a 1-hour write costs $10.00.
  • Cached input is Anthropic's cache hits and refreshes price.
  • Context window from the model page.
Anthropic$5.00$0.50$25.00$2.50$12.501M$0.075Pricing pagechecked 2026-11-12
Claude Opus 4.6
Notes
  • Cache write: $6.25.
  • Cache write is Anthropic's 5-minute write price; a 1-hour write costs $10.00.
  • Cached input is Anthropic's cache hits and refreshes price.
  • Context window from the model page.
Anthropic$5.00$0.50$25.00$2.50$12.501M$0.075Pricing pagechecked 2026-11-12
Claude Opus 4.5
Notes
  • Cache write: $6.25.
  • Cache write is Anthropic's 5-minute write price; a 1-hour write costs $10.00.
  • Cached input is Anthropic's cache hits and refreshes price.
  • Context window from the model page.
Anthropic$5.00$0.50$25.00$2.50$12.50200K$0.075Pricing pagechecked 2026-11-12
Claude Sonnet 5
Notes
  • Cache write: $2.50.
  • Cache write is Anthropic's 5-minute write price; a 1-hour write costs $4.00.
  • Cached input is Anthropic's cache hits and refreshes price.
  • Context window from the model page.
Anthropic$2.00$0.20$10.00$1.00$5.001M$0.03Pricing pagechecked 2026-11-12
Claude Sonnet 4.6
Notes
  • Cache write: $3.75.
  • Cache write is Anthropic's 5-minute write price; a 1-hour write costs $6.00.
  • Cached input is Anthropic's cache hits and refreshes price.
  • Context window from the model page.
Anthropic$3.00$0.30$15.00$1.50$7.501M$0.045Pricing pagechecked 2026-11-12
Claude Sonnet 4.5
Notes
  • Cache write: $3.75.
  • Cache write is Anthropic's 5-minute write price; a 1-hour write costs $6.00.
  • Cached input is Anthropic's cache hits and refreshes price.
  • Context window from the model page.
Anthropic$3.00$0.30$15.00$1.50$7.50200K$0.045Pricing pagechecked 2026-11-12
Claude Haiku 4.5
Notes
  • Cache write: $1.25.
  • Cache write is Anthropic's 5-minute write price; a 1-hour write costs $2.00.
  • Cached input is Anthropic's cache hits and refreshes price.
  • Context window from the model page.
Anthropic$1.00$0.10$5.00$0.50$2.50200K$0.015Pricing pagechecked 2026-11-12
Gemini 4 Carbongemini-4-carbon
Notes
  • From 2026-12-01: input $4.00, cached input $0.20, output $20.00, batch input $2.00, batch output $10.00.
  • Text prices. Audio input and context cache storage (per hour) are priced separately on the page.
  • Implicit caching only: cache hits are billed at the cached price, with no separate write charge.
  • Context window from the model page.
Google$2.00$0.10$10.00$1.00$5.001.05M$0.03Pricing pagechecked 2026-11-12
Gemini 4 Argongemini-4-argon
Notes
  • From 2026-12-01: input $4.00, cached input $0.20, output $20.00, batch input $2.00, batch output $10.00.
  • Text prices. Audio input and context cache storage (per hour) are priced separately on the page.
  • Implicit caching only: cache hits are billed at the cached price, with no separate write charge.
  • Context window from the model page.
Google$2.00$0.10$10.00$1.00$5.001.05M$0.03Pricing pagechecked 2026-11-12
Gemini 4 Hydrogengemini-4-hydrogen
Notes
  • Text prices. Audio input and context cache storage (per hour) are priced separately on the page.
  • Implicit caching only: cache hits are billed at the cached price, with no separate write charge.
  • Context window from the model page.
Google$1.50$0.075$7.50$0.75$3.751.05M$0.0225Pricing pagechecked 2026-11-12
Gemini 3.8 Flashgemini-3.8-flash
Notes
  • From 2027-01-01: input $1.50, cached input $0.15, output $7.50, batch input $0.75, batch output $3.75.
  • Text prices. Audio input and context cache storage (per hour) are priced separately on the page.
  • Context window from the model page.
Google$0.75$0.075$3.75$0.375$1.8751.05M$0.01125Pricing pagechecked 2026-11-12
Gemini 3.6 Flashgemini-3.6-flash
Notes
  • From 2027-01-01: input $1.50, cached input $0.15, output $7.50, batch input $0.75, batch output $3.75.
  • Text prices. Audio input and context cache storage (per hour) are priced separately on the page.
  • Context window from the model page.
Google$0.75$0.075$3.75$0.375$1.8751.05M$0.01125Pricing pagechecked 2026-11-12
Gemini 3.5 Flash-Litegemini-3.5-flash-lite
Notes
  • Text prices. Audio input and context cache storage (per hour) are priced separately on the page.
  • Context window from the model page.
Google$0.30$0.03$2.50$0.15$1.251.05M$0.0055Pricing pagechecked 2026-11-12
Gemini 3.1 Flash-Litegemini-3.1-flash-lite
Notes
  • Text prices. Audio input and context cache storage (per hour) are priced separately on the page.
  • Context window from the model page.
Google$0.25$0.025$1.50$0.125$0.751.05M$0.004Pricing pagechecked 2026-11-12
Gemini 3.1 Pro Previewgemini-3.1-pro-preview
Notes
  • Prompts over 200,000 tokens: input $4.00, cached input $0.40, output $18.00, batch input $2.00, batch output $9.00.
  • Text prices. Audio input and context cache storage (per hour) are priced separately on the page.
  • Context window from the model page.
Google$2.00$0.20$12.00$1.00$6.001.05M$0.032Pricing pagechecked 2026-11-12
Gemini 3 Flash Previewgemini-3-flash-preview
Notes
  • Text prices. Audio input and context cache storage (per hour) are priced separately on the page.
  • Context window from the model page.
Google$0.50$0.05$3.00$0.25$1.501.05M$0.008Pricing pagechecked 2026-11-12
Gemini 2.5 Progemini-2.5-pro
Notes
  • Listed price ends after 2026-11-30.
  • Prompts over 200,000 tokens: input $2.50, cached input $0.25, output $15.00, batch input $1.25, batch output $7.50.
  • Text prices. Audio input and context cache storage (per hour) are priced separately on the page.
  • Google lists this model as deprecated. From December 1, 2026, requests to it are routed to Gemini 3.8 Flash (gemini-3.8-flash) and billed at that model's rates.
  • Context window from the model page.
Google$1.25$0.125$10.00$0.625$5.001.05M$0.0225Pricing pagechecked 2026-11-12
Gemini 2.5 Flashgemini-2.5-flash
Notes
  • Listed price ends after 2026-11-30.
  • Text prices. Audio input and context cache storage (per hour) are priced separately on the page.
  • Google lists this model as deprecated. From December 1, 2026, requests to it are routed to Gemini 3.5 Flash-Lite (gemini-3.5-flash-lite) and billed at that model's rates.
  • Context window from the model page.
Google$0.30$0.03$2.50$0.15$1.251.05M$0.0055Pricing pagechecked 2026-11-12
Gemini 2.5 Flash-Litegemini-2.5-flash-lite
Notes
  • Text prices. Audio input and context cache storage (per hour) are priced separately on the page.
  • Context window from the model page.
Google$0.10$0.01$0.40$0.05$0.201.05M$0.0014Pricing pagechecked 2026-11-12
Mistral Large 4
Notes
  • Mistral says Batch is 50% off, but lists no per-model batch price, so none is shown.
  • Context window from the model page.
Mistral$0.70$0.07$2.10n/an/a1M$0.0091Pricing pagechecked 2026-11-12
Mistral Small 4
Notes
  • Mistral says Batch is 50% off, but lists no per-model batch price, so none is shown.
  • Context window from the model page.
Mistral$0.15$0.015$0.60n/an/a256K$0.0021Pricing pagechecked 2026-11-12
Ministral 3 14B
Notes
  • Mistral says Batch is 50% off, but lists no per-model batch price, so none is shown.
  • Context window from the model page.
Mistral$0.20$0.02$0.20n/an/a256K$0.0022Pricing pagechecked 2026-11-12
Ministral 3 8B
Notes
  • Mistral says Batch is 50% off, but lists no per-model batch price, so none is shown.
  • Context window from the model page.
Mistral$0.15$0.015$0.15n/an/a256K$0.00165Pricing pagechecked 2026-11-12
Ministral 3 3B
Notes
  • Mistral says Batch is 50% off, but lists no per-model batch price, so none is shown.
  • Context window from the model page.
Mistral$0.10$0.01$0.10n/an/a256K$0.0011Pricing pagechecked 2026-11-12
Codestral
Notes
  • Mistral says Batch is 50% off, but lists no per-model batch price, so none is shown.
  • Context window from the model page.
Mistral$0.30$0.03$0.90n/an/a128K$0.0039Pricing pagechecked 2026-11-12

How to read the table

Prices are US dollars per 1 million tokens at standard processing, as each vendor lists them. Batch is the vendor's listed price for its batch API, where one is published. n/a means the vendor lists no such price for that model.

Your request = ( 10,000 × input + 1,000 × output ) ÷ 1,000,000, at standard prices with no caching.

Cached input is what a cache hit costs. Some vendors also charge to write the cache; that price is in the model's notes. Prices that change on a set date, or that depend on prompt length, are in the notes too, and the table switches to a new price on the day it starts.

Sources and privacy

Where these prices come from

  • Every price is copied from the vendor's own pricing page, linked on its row, on the date shown. We list no price we could not find there.
  • Context windows come from each vendor's model pages. Where a vendor publishes none, the table says "not listed".
  • The page runs in your browser and loads its price list from this site. Search, filters and your request sizes are never sent anywhere; a shared link keeps them after the #.
  • To compare cost per conversation with prompt caching and routing, use our AI cost-per-conversation calculator.

Want us to look at the real thing?

Book a free 30-minute consult: we'll look at your bill with you, no obligation and no sales pitch. Or check your own cloud bill first with our free Cloud Bill Gotchas Checker.

Book a free 30-minute consult