Context Window Comparison
42 LLMs ranked by context window size. Larger context means more text can be processed in a single conversation — crucial for analysing long documents, codebases, and research papers.
11
1M+ tokens
12
200K–999K
19
32K–199K
0
<32K tokens
Best Quality by Context Size
Context Size Ranking
Full Table
| # | Model | Context | Quality |
|---|---|---|---|
| 1 | Gemini 2.5 Pro Google | 1.0M | 83.0 |
| 2 | Gemini 2.0 Flash Google | 1.0M | 81.0 |
| 3 | Gemini 2.5 Flash Google | 1.0M | 78.0 |
| 4 | Gemini 2.5 Flash Lite Google | 1.0M | 78.0 |
| 5 | Gemini 2.0 Flash Lite Google | 1.0M | 76.0 |
| 6 | Llama 4 Maverick OSS Meta | 1.0M | 75.0 |
| 7 | GPT-4.1 OpenAI | 1.0M | 77.0 |
| 8 | GPT-4.1 Nano OpenAI | 1.0M | 75.0 |
| 9 | Claude Opus 4.6 Anthropic | 1M | 89.0 |
| 10 | Claude Sonnet 4.6 Anthropic | 1M | 86.0 |
| 11 | Claude Sonnet 4 Anthropic | 1M | 79.0 |
| 12 | GPT-5.2 OpenAI | 400K | 90.0 |
| 13 | GPT-5 OpenAI | 400K | 87.0 |
| 14 | GPT-5 Nano OpenAI | 400K | 78.0 |
| 15 | Nova Pro 1.0 Amazon | 300K | 78.0 |
| 16 | Nova Lite 1.0 Amazon | 300K | 72.0 |
| 17 | Grok 4 xAI | 256K | 88.0 |
| 18 | Command A OSS Cohere | 256K | 82.0 |
| 19 | O4 Mini OpenAI | 200K | 90.0 |
| 20 | O3 OpenAI | 200K | 88.0 |
| 21 | O3 Pro OpenAI | 200K | 88.0 |
| 22 | Claude Opus 4 Anthropic | 200K | 84.0 |
| 23 | Claude 3.5 Haiku Anthropic | 200K | 82.0 |
| 24 | DeepSeek V3.2 OSS DeepSeek | 163.8K | 86.0 |
| 25 | DeepSeek V3 OSS DeepSeek | 163.8K | 76.0 |
| 26 | Qwen3 235B A22B OSS Alibaba | 131.1K | 87.0 |
| 27 | Grok 3 Beta xAI | 131.1K | 85.0 |
| 28 | Llama 3.3 70B Instruct OSS Meta | 131.1K | 79.0 |
| 29 | QwQ 32B OSS Alibaba | 131.1K | 78.0 |
| 30 | Mistral Nemo OSS Mistral | 131.1K | 72.0 |
| 31 | Mistral Large OSS Mistral | 128K | 86.0 |
| 32 | GPT-4o-mini OpenAI | 128K | 80.0 |
| 33 | Command R+ (08-2024) OSS Cohere | 128K | 79.0 |
| 34 | Mistral Small 3.1 24B OSS Mistral | 128K | 76.0 |
| 35 | GPT-4o (extended) OpenAI | 128K | 75.0 |
| 36 | Command R (08-2024) OSS Cohere | 128K | 73.0 |
| 37 | Nova Micro 1.0 Amazon | 128K | 68.0 |
| 38 | Command R7B (12-2024) OSS Cohere | 128K | 65.0 |
| 39 | Sonar Perplexity | 127.1K | 74.0 |
| 40 | Reka Flash 3 Reka | 65.5K | 74.0 |
| 41 | DeepSeek R1 OSS DeepSeek | 64K | 85.0 |
| 42 | Qwen2.5 72B Instruct OSS Alibaba | 32.8K | 71.0 |
What is a context window?
The context window is the maximum amount of text a model can process in a single request — including both your input and its response. It's measured in tokens (roughly 0.75 words per token).
32K tokens
~24,000 words / ~96 pages. Enough for a long article or short story.
200K tokens
~150,000 words / ~600 pages. Enough for a full novel or large codebase.
1M+ tokens
~750,000+ words / ~3,000+ pages. Entire book series, repos, or document collections.