Claude Fable 5$22.000/MClaude Opus 5$11.000/MClaude Opus 4.8$11.000/MClaude Opus 4.7$11.000/MClaude Opus 4.6$11.000/MClaude Opus 4.5$33.000/MClaude Sonnet 3.7$6.600/MClaude Opus 3$33.000/MClaude 2.1$12.800/MClaude 2$12.800/MGPT-5.6 Sol$12.500/MGPT-5.6 Terra$5.000/MGPT-5.5$12.500/MGPT-5.2$5.425/MGPT-5.2-Codex$5.425/MGPT-5$3.875/MGPT-4.5$97.500/MGPT-4 Turbo Preview$16.000/MGPT-4$39.000/MGPT-4-32k$78.000/Mo3$19.000/Mo3-mini$2.090/Mo4-mini$2.090/Mo1$28.500/Mo1-mini$5.700/Mo1-preview$28.500/MGemini 3.5 Pro$5.000/MGemini 3.1 Pro$5.000/MGemini 3 Pro$5.000/MGemini 2.5 Pro$3.875/MClaude Fable 5$22.000/MClaude Opus 5$11.000/MClaude Opus 4.8$11.000/MClaude Opus 4.7$11.000/MClaude Opus 4.6$11.000/MClaude Opus 4.5$33.000/MClaude Sonnet 3.7$6.600/MClaude Opus 3$33.000/MClaude 2.1$12.800/MClaude 2$12.800/MGPT-5.6 Sol$12.500/MGPT-5.6 Terra$5.000/MGPT-5.5$12.500/MGPT-5.2$5.425/MGPT-5.2-Codex$5.425/MGPT-5$3.875/MGPT-4.5$97.500/MGPT-4 Turbo Preview$16.000/MGPT-4$39.000/MGPT-4-32k$78.000/Mo3$19.000/Mo3-mini$2.090/Mo4-mini$2.090/Mo1$28.500/Mo1-mini$5.700/Mo1-preview$28.500/MGemini 3.5 Pro$5.000/MGemini 3.1 Pro$5.000/MGemini 3 Pro$5.000/MGemini 2.5 Pro$3.875/M

Longest Context Window AI Models

Ranked from provider-published pricing · Prices checked 12 August 2026

The context window is the hard ceiling on how much text a model can consider in one request: prompt, retrieved documents, conversation history and its own output all share it. This list ranks the catalogue by that ceiling, largest first.

Prices are shown beside each window so the tradeoff is visible, the largest windows are not always the most expensive models.

The ranking

top 20 of 271

Gemini 1.5 Pro from Google leads this ranking at 2,000,000 tokens. The median across the 271 qualifying models is 128,000 tokens.

#ModelProviderMaximum Context WindowInput /1MOutput /1MBlended 70/30
1Gemini 1.5 ProGoogle2,000,000 tokens$1.25$5.00$2.38
2GPT-5.6 SolOpenAI1,050,000 tokens$5.00$30.00$12.50
3GPT-5.6 TerraOpenAI1,050,000 tokens$2.00$12.00$5.00
4GPT-5.6 LunaOpenAI1,050,000 tokens$0.200$1.20$0.500
5GPT-5.5OpenAI1,050,000 tokens$5.00$30.00$12.50
6Gemini 3.6 FlashGoogle1,050,000 tokens$1.50$7.50$3.30
7Muse Spark 1.2Meta1,049,000 tokens$1.25$4.25$2.15
8Kimi K3Moonshot AI1,049,000 tokens$3.00$15.00$6.60
9Claude Fable 5Anthropic1,000,000 tokens$10.00$50.00$22.00
10Claude Opus 5Anthropic1,000,000 tokens$5.00$25.00$11.00
11Claude Sonnet 5Anthropic1,000,000 tokens$2.00$10.00$4.40
12Claude Opus 4.8Anthropic1,000,000 tokens$5.00$25.00$11.00
13Claude Opus 4.7Anthropic1,000,000 tokens$5.00$25.00$11.00
14Claude Opus 4.6Anthropic1,000,000 tokens$5.00$25.00$11.00
15Claude Sonnet 4.6Anthropic1,000,000 tokens$3.00$15.00$6.60
16GPT-4.1OpenAI1,000,000 tokens$2.00$8.00$3.80
17GPT-4.1 miniOpenAI1,000,000 tokens$0.400$1.60$0.760
18GPT-4.1 nanoOpenAI1,000,000 tokens$0.100$0.400$0.190
19Gemini 3.5 ProGoogle1,000,000 tokens$2.00$12.00$5.00
20Gemini 3.5 FlashGoogle1,000,000 tokens$1.50$9.00$3.75

How this list is built

  • Ranked by maximum context window, largest first.
  • Excludes embed models.
  • Includes models marked available or preview; retired and deprecated SKUs are excluded.
  • Prices are the rates each provider publishes, not estimates. How the blended rate is calculated.

Frequently asked questions

What is a context window?

It is the maximum number of tokens a model can process in a single request, covering everything you send plus everything it generates. Exceeding it forces you to truncate, summarise or chunk the input. As a rough guide, one million tokens is somewhere around 750,000 words.

Do I need a 1M-token context window?

Rarely. Most production systems retrieve the handful of relevant passages instead of sending an entire corpus, which is both cheaper and usually more accurate, since models can lose track of details buried in a very long context. Large windows matter most for whole-codebase reasoning and long document analysis where chunking loses the thread.

Related rankings

Price it for your workload

Rankings are recomputed on every deploy from the catalogue. Prices checked 12 August 2026.