Context windows, ranked
A model's context window is the maximum number of tokens it can consider in a single request — your prompt, any documents you attach, and for most APIs the response too. This table ranks all 338 models by context length, with a plain-terms estimate of what actually fits. Not sure how many tokens your text is? Use the token calculator.
- 10M
- Largest context
- 75
- Models at 1M+
- 203K
- Median context
- 338
- Models
ModelContextTokens≈ Pages
Llama 4 ScoutMeta AI10M10,000,000≈ 15,00010M · 10,000,000 tokens · ≈ 15,000 pages
Auto RouterOpenRouter2M2,000,000≈ 3,0002M · 2,000,000 tokens · ≈ 3,000 pages
Grok 4.20xAI2M2,000,000≈ 3,0002M · 2,000,000 tokens · ≈ 3,000 pages- Grok 4.20 Multi-AgentxAI2M2,000,000≈ 3,0002M · 2,000,000 tokens · ≈ 3,000 pages
- Pareto Code RouterOpenRouter2M2,000,000≈ 3,0002M · 2,000,000 tokens · ≈ 3,000 pages
GPT-5.4OpenAI1.1M1,050,000≈ 1,5751.1M · 1,050,000 tokens · ≈ 1,575 pages- GPT-5.4 ProOpenAI1.1M1,050,000≈ 1,5751.1M · 1,050,000 tokens · ≈ 1,575 pages
- GPT-5.5OpenAI1.1M1,050,000≈ 1,5751.1M · 1,050,000 tokens · ≈ 1,575 pages
- GPT-5.5 ProOpenAI1.1M1,050,000≈ 1,5751.1M · 1,050,000 tokens · ≈ 1,575 pages
- OpenAI GPT LatestOpenAI1.1M1,050,000≈ 1,5751.1M · 1,050,000 tokens · ≈ 1,575 pages
Gemini 3.1 Pro Preview Custom ToolsGoogle1.0M1,048,756≈ 1,5731.0M · 1,048,756 tokens · ≈ 1,573 pages
DeepSeek V4 FlashDeepSeek1.0M1,048,576≈ 1,5731.0M · 1,048,576 tokens · ≈ 1,573 pages- DeepSeek V4 ProDeepSeek1.0M1,048,576≈ 1,5731.0M · 1,048,576 tokens · ≈ 1,573 pages
- Gemini 2.5 FlashGoogle1.0M1,048,576≈ 1,5731.0M · 1,048,576 tokens · ≈ 1,573 pages
- Gemini 2.5 Flash LiteGoogle1.0M1,048,576≈ 1,5731.0M · 1,048,576 tokens · ≈ 1,573 pages
- Gemini 2.5 Flash Lite Preview 09-2025Google1.0M1,048,576≈ 1,5731.0M · 1,048,576 tokens · ≈ 1,573 pages
- Gemini 2.5 ProGoogle1.0M1,048,576≈ 1,5731.0M · 1,048,576 tokens · ≈ 1,573 pages
- Gemini 2.5 Pro Preview 05-06Google1.0M1,048,576≈ 1,5731.0M · 1,048,576 tokens · ≈ 1,573 pages
- Gemini 2.5 Pro Preview 06-05Google1.0M1,048,576≈ 1,5731.0M · 1,048,576 tokens · ≈ 1,573 pages
- Gemini 3 Flash PreviewGoogle1.0M1,048,576≈ 1,5731.0M · 1,048,576 tokens · ≈ 1,573 pages
- Gemini 3.1 Flash LiteGoogle1.0M1,048,576≈ 1,5731.0M · 1,048,576 tokens · ≈ 1,573 pages
- Gemini 3.1 Flash Lite PreviewGoogle1.0M1,048,576≈ 1,5731.0M · 1,048,576 tokens · ≈ 1,573 pages
- Gemini 3.1 Pro PreviewGoogle1.0M1,048,576≈ 1,5731.0M · 1,048,576 tokens · ≈ 1,573 pages
- Gemini 3.5 FlashGoogle1.0M1,048,576≈ 1,5731.0M · 1,048,576 tokens · ≈ 1,573 pages
GLM 5.2Z.ai1.0M1,048,576≈ 1,5731.0M · 1,048,576 tokens · ≈ 1,573 pages- Google Gemini Flash LatestGoogle1.0M1,048,576≈ 1,5731.0M · 1,048,576 tokens · ≈ 1,573 pages
- Google Gemini Pro LatestGoogle1.0M1,048,576≈ 1,5731.0M · 1,048,576 tokens · ≈ 1,573 pages
- Llama 4 MaverickMeta AI1.0M1,048,576≈ 1,5731.0M · 1,048,576 tokens · ≈ 1,573 pages
- Lyria 3 Clip PreviewGoogleFree1.0M1,048,576≈ 1,5731.0M · 1,048,576 tokens · ≈ 1,573 pages
- Lyria 3 Pro PreviewGoogleFree1.0M1,048,576≈ 1,5731.0M · 1,048,576 tokens · ≈ 1,573 pages
MiMo-V2.5Xiaomi1.0M1,048,576≈ 1,5731.0M · 1,048,576 tokens · ≈ 1,573 pages- MiMo-V2.5-ProXiaomi1.0M1,048,576≈ 1,5731.0M · 1,048,576 tokens · ≈ 1,573 pages
MiniMax M3MiniMax1.0M1,048,576≈ 1,5731.0M · 1,048,576 tokens · ≈ 1,573 pages
Qwen3 Coder 480B A35BQwen1.0M1,048,576≈ 1,5731.0M · 1,048,576 tokens · ≈ 1,573 pages- Qwen3 Coder 480B A35B (free)QwenFree1.0M1,048,576≈ 1,5731.0M · 1,048,576 tokens · ≈ 1,573 pages
- GPT-4.1OpenAI1.0M1,047,576≈ 1,5711.0M · 1,047,576 tokens · ≈ 1,571 pages
- GPT-4.1 MiniOpenAI1.0M1,047,576≈ 1,5711.0M · 1,047,576 tokens · ≈ 1,571 pages
- GPT-4.1 NanoOpenAI1.0M1,047,576≈ 1,5711.0M · 1,047,576 tokens · ≈ 1,571 pages
Palmyra X5Writer1.0M1,040,000≈ 1,5601.0M · 1,040,000 tokens · ≈ 1,560 pages- MiniMax-01MiniMax1.0M1,000,192≈ 1,5001.0M · 1,000,192 tokens · ≈ 1,500 pages
Anthropic Claude Sonnet LatestAnthropic1M1,000,000≈ 1,5001M · 1,000,000 tokens · ≈ 1,500 pages- Claude Fable 5Anthropic1M1,000,000≈ 1,5001M · 1,000,000 tokens · ≈ 1,500 pages
- Claude Fable LatestAnthropic1M1,000,000≈ 1,5001M · 1,000,000 tokens · ≈ 1,500 pages
- Claude Opus 4.6Anthropic1M1,000,000≈ 1,5001M · 1,000,000 tokens · ≈ 1,500 pages
- Claude Opus 4.7Anthropic1M1,000,000≈ 1,5001M · 1,000,000 tokens · ≈ 1,500 pages
- Claude Opus 4.7 (Fast)Anthropic1M1,000,000≈ 1,5001M · 1,000,000 tokens · ≈ 1,500 pages
- Claude Opus 4.8Anthropic1M1,000,000≈ 1,5001M · 1,000,000 tokens · ≈ 1,500 pages
- Claude Opus 4.8 (Fast)Anthropic1M1,000,000≈ 1,5001M · 1,000,000 tokens · ≈ 1,500 pages
- Claude Opus LatestAnthropic1M1,000,000≈ 1,5001M · 1,000,000 tokens · ≈ 1,500 pages
- Claude Sonnet 4Anthropic1M1,000,000≈ 1,5001M · 1,000,000 tokens · ≈ 1,500 pages
- Claude Sonnet 4.5Anthropic1M1,000,000≈ 1,5001M · 1,000,000 tokens · ≈ 1,500 pages
- Claude Sonnet 4.6Anthropic1M1,000,000≈ 1,5001M · 1,000,000 tokens · ≈ 1,500 pages
- Claude Sonnet 5Anthropic1M1,000,000≈ 1,5001M · 1,000,000 tokens · ≈ 1,500 pages
Fugu UltraSakana AI1M1,000,000≈ 1,5001M · 1,000,000 tokens · ≈ 1,500 pages- FusionOpenRouter1M1,000,000≈ 1,5001M · 1,000,000 tokens · ≈ 1,500 pages
- Grok 4.3xAI1M1,000,000≈ 1,5001M · 1,000,000 tokens · ≈ 1,500 pages
- MiniMax M1MiniMax1M1,000,000≈ 1,5001M · 1,000,000 tokens · ≈ 1,500 pages
Nemotron 3 SuperNVIDIA1M1,000,000≈ 1,5001M · 1,000,000 tokens · ≈ 1,500 pages- Nemotron 3 Super (free)NVIDIAFree1M1,000,000≈ 1,5001M · 1,000,000 tokens · ≈ 1,500 pages
- Nemotron 3 UltraNVIDIA1M1,000,000≈ 1,5001M · 1,000,000 tokens · ≈ 1,500 pages
- Nemotron 3 Ultra (free)NVIDIAFree1M1,000,000≈ 1,5001M · 1,000,000 tokens · ≈ 1,500 pages
Nova 2 LiteAmazon1M1,000,000≈ 1,5001M · 1,000,000 tokens · ≈ 1,500 pages- Nova Premier 1.0Amazon1M1,000,000≈ 1,5001M · 1,000,000 tokens · ≈ 1,500 pages
- Qwen Plus 0728Qwen1M1,000,000≈ 1,5001M · 1,000,000 tokens · ≈ 1,500 pages
- Qwen Plus 0728 (thinking)Qwen1M1,000,000≈ 1,5001M · 1,000,000 tokens · ≈ 1,500 pages
- Qwen-PlusQwen1M1,000,000≈ 1,5001M · 1,000,000 tokens · ≈ 1,500 pages
- Qwen3 Coder FlashQwen1M1,000,000≈ 1,5001M · 1,000,000 tokens · ≈ 1,500 pages
- Qwen3 Coder PlusQwen1M1,000,000≈ 1,5001M · 1,000,000 tokens · ≈ 1,500 pages
- Qwen3.5 Plus 2026-02-15Qwen1M1,000,000≈ 1,5001M · 1,000,000 tokens · ≈ 1,500 pages
- Qwen3.5 Plus 2026-04-20Qwen1M1,000,000≈ 1,5001M · 1,000,000 tokens · ≈ 1,500 pages
- Qwen3.5-FlashQwen1M1,000,000≈ 1,5001M · 1,000,000 tokens · ≈ 1,500 pages
- Qwen3.6 FlashQwen1M1,000,000≈ 1,5001M · 1,000,000 tokens · ≈ 1,500 pages
- Qwen3.6 PlusQwen1M1,000,000≈ 1,5001M · 1,000,000 tokens · ≈ 1,500 pages
- Qwen3.7 MaxQwen1M1,000,000≈ 1,5001M · 1,000,000 tokens · ≈ 1,500 pages
- Qwen3.7 PlusQwen1M1,000,000≈ 1,5001M · 1,000,000 tokens · ≈ 1,500 pages
- GPT Chat LatestOpenAI400K400,000≈ 600400K · 400,000 tokens · ≈ 600 pages
- GPT-5OpenAI400K400,000≈ 600400K · 400,000 tokens · ≈ 600 pages
- GPT-5 CodexOpenAI400K400,000≈ 600400K · 400,000 tokens · ≈ 600 pages
- GPT-5 ImageOpenAI400K400,000≈ 600400K · 400,000 tokens · ≈ 600 pages
- GPT-5 Image MiniOpenAI400K400,000≈ 600400K · 400,000 tokens · ≈ 600 pages
- GPT-5 MiniOpenAI400K400,000≈ 600400K · 400,000 tokens · ≈ 600 pages
- GPT-5 NanoOpenAI400K400,000≈ 600400K · 400,000 tokens · ≈ 600 pages
- GPT-5 ProOpenAI400K400,000≈ 600400K · 400,000 tokens · ≈ 600 pages
- GPT-5.1OpenAI400K400,000≈ 600400K · 400,000 tokens · ≈ 600 pages
- GPT-5.1-CodexOpenAI400K400,000≈ 600400K · 400,000 tokens · ≈ 600 pages
- GPT-5.1-Codex-MaxOpenAI400K400,000≈ 600400K · 400,000 tokens · ≈ 600 pages
- GPT-5.1-Codex-MiniOpenAI400K400,000≈ 600400K · 400,000 tokens · ≈ 600 pages
- GPT-5.2OpenAI400K400,000≈ 600400K · 400,000 tokens · ≈ 600 pages
- GPT-5.2 ProOpenAI400K400,000≈ 600400K · 400,000 tokens · ≈ 600 pages
- GPT-5.2-CodexOpenAI400K400,000≈ 600400K · 400,000 tokens · ≈ 600 pages
- GPT-5.3-CodexOpenAI400K400,000≈ 600400K · 400,000 tokens · ≈ 600 pages
- GPT-5.4 MiniOpenAI400K400,000≈ 600400K · 400,000 tokens · ≈ 600 pages
- GPT-5.4 NanoOpenAI400K400,000≈ 600400K · 400,000 tokens · ≈ 600 pages
- OpenAI GPT Mini LatestOpenAI400K400,000≈ 600400K · 400,000 tokens · ≈ 600 pages
- Nova Lite 1.0Amazon300K300,000≈ 450300K · 300,000 tokens · ≈ 450 pages
- Nova Pro 1.0Amazon300K300,000≈ 450300K · 300,000 tokens · ≈ 450 pages
- GPT-5.4 Image 2OpenAI272K272,000≈ 408272K · 272,000 tokens · ≈ 408 pages
Devstral 2 2512Mistral AI262K262,144≈ 393262K · 262,144 tokens · ≈ 393 pages- Gemma 4 26B A4B Google262K262,144≈ 393262K · 262,144 tokens · ≈ 393 pages
- Gemma 4 26B A4B (free)GoogleFree262K262,144≈ 393262K · 262,144 tokens · ≈ 393 pages
- Gemma 4 31BGoogle262K262,144≈ 393262K · 262,144 tokens · ≈ 393 pages
- Gemma 4 31B (free)GoogleFree262K262,144≈ 393262K · 262,144 tokens · ≈ 393 pages
- GLM 5 TurboZ.ai262K262,144≈ 393262K · 262,144 tokens · ≈ 393 pages
Hy3 previewTencent262K262,144≈ 393262K · 262,144 tokens · ≈ 393 pages
Kimi K2 0905Moonshot AI262K262,144≈ 393262K · 262,144 tokens · ≈ 393 pages- Kimi K2 ThinkingMoonshot AI262K262,144≈ 393262K · 262,144 tokens · ≈ 393 pages
- Kimi K2.5Moonshot AI262K262,144≈ 393262K · 262,144 tokens · ≈ 393 pages
- Kimi K2.6Moonshot AI262K262,144≈ 393262K · 262,144 tokens · ≈ 393 pages
- Kimi K2.7 CodeMoonshot AI262K262,144≈ 393262K · 262,144 tokens · ≈ 393 pages
Laguna M.1poolside262K262,144≈ 393262K · 262,144 tokens · ≈ 393 pages- Laguna M.1 (free)poolsideFree262K262,144≈ 393262K · 262,144 tokens · ≈ 393 pages
- Laguna XS.2poolside262K262,144≈ 393262K · 262,144 tokens · ≈ 393 pages
- Laguna XS.2 (free)poolsideFree262K262,144≈ 393262K · 262,144 tokens · ≈ 393 pages
Ling-2.6-1TInclusionAI262K262,144≈ 393262K · 262,144 tokens · ≈ 393 pages- Ling-2.6-flashInclusionAI262K262,144≈ 393262K · 262,144 tokens · ≈ 393 pages
- Ministral 3 14B 2512Mistral AI262K262,144≈ 393262K · 262,144 tokens · ≈ 393 pages
- Ministral 3 8B 2512Mistral AI262K262,144≈ 393262K · 262,144 tokens · ≈ 393 pages
- Mistral Large 3 2512Mistral AI262K262,144≈ 393262K · 262,144 tokens · ≈ 393 pages
- Mistral Medium 3.5Mistral AI262K262,144≈ 393262K · 262,144 tokens · ≈ 393 pages
- Mistral Small 4Mistral AI262K262,144≈ 393262K · 262,144 tokens · ≈ 393 pages
- MoonshotAI Kimi LatestMoonshot AI262K262,144≈ 393262K · 262,144 tokens · ≈ 393 pages
Morph V3 LargeMorph262K262,144≈ 393262K · 262,144 tokens · ≈ 393 pages- Nemotron 3 Nano 30B A3BNVIDIA262K262,144≈ 393262K · 262,144 tokens · ≈ 393 pages
Nex-N2-ProNex262K262,144≈ 393262K · 262,144 tokens · ≈ 393 pages- Qwen3 235B A22B Instruct 2507Qwen262K262,144≈ 393262K · 262,144 tokens · ≈ 393 pages
- Qwen3 235B A22B Thinking 2507Qwen262K262,144≈ 393262K · 262,144 tokens · ≈ 393 pages
- Qwen3 Coder NextQwen262K262,144≈ 393262K · 262,144 tokens · ≈ 393 pages
- Qwen3 MaxQwen262K262,144≈ 393262K · 262,144 tokens · ≈ 393 pages
- Qwen3 Max ThinkingQwen262K262,144≈ 393262K · 262,144 tokens · ≈ 393 pages
- Qwen3 Next 80B A3B InstructQwen262K262,144≈ 393262K · 262,144 tokens · ≈ 393 pages
- Qwen3 Next 80B A3B Instruct (free)QwenFree262K262,144≈ 393262K · 262,144 tokens · ≈ 393 pages
- Qwen3 Next 80B A3B ThinkingQwen262K262,144≈ 393262K · 262,144 tokens · ≈ 393 pages
- Qwen3 VL 235B A22B InstructQwen262K262,144≈ 393262K · 262,144 tokens · ≈ 393 pages
- Qwen3 VL 30B A3B InstructQwen262K262,144≈ 393262K · 262,144 tokens · ≈ 393 pages
- Qwen3 VL 32B InstructQwen262K262,144≈ 393262K · 262,144 tokens · ≈ 393 pages
- Qwen3.5-122B-A10BQwen262K262,144≈ 393262K · 262,144 tokens · ≈ 393 pages
- Qwen3.5-27BQwen262K262,144≈ 393262K · 262,144 tokens · ≈ 393 pages
- Qwen3.5-35B-A3BQwen262K262,144≈ 393262K · 262,144 tokens · ≈ 393 pages
- Qwen3.5-9BQwen262K262,144≈ 393262K · 262,144 tokens · ≈ 393 pages
- Qwen3.6 27BQwen262K262,144≈ 393262K · 262,144 tokens · ≈ 393 pages
- Qwen3.6 35B A3BQwen262K262,144≈ 393262K · 262,144 tokens · ≈ 393 pages
- Qwen3.6 Max PreviewQwen262K262,144≈ 393262K · 262,144 tokens · ≈ 393 pages
- Ring-2.6-1TInclusionAI262K262,144≈ 393262K · 262,144 tokens · ≈ 393 pages
Seed 1.6ByteDance Seed262K262,144≈ 393262K · 262,144 tokens · ≈ 393 pages- Seed 1.6 FlashByteDance Seed262K262,144≈ 393262K · 262,144 tokens · ≈ 393 pages
- Seed-2.0-LiteByteDance Seed262K262,144≈ 393262K · 262,144 tokens · ≈ 393 pages
- Seed-2.0-MiniByteDance Seed262K262,144≈ 393262K · 262,144 tokens · ≈ 393 pages
Step 3.5 FlashStepFun262K262,144≈ 393262K · 262,144 tokens · ≈ 393 pages
Trinity Large ThinkingArcee AI262K262,144≈ 393262K · 262,144 tokens · ≈ 393 pages- Codestral 2508Mistral AI256K256,000≈ 384256K · 256,000 tokens · ≈ 384 pages
Command ACohere256K256,000≈ 384256K · 256,000 tokens · ≈ 384 pages- Grok Build 0.1xAI256K256,000≈ 384256K · 256,000 tokens · ≈ 384 pages
Jamba Large 1.7AI21 Labs256K256,000≈ 384256K · 256,000 tokens · ≈ 384 pages
KAT-Coder-Pro V2Kwaipilot256K256,000≈ 384256K · 256,000 tokens · ≈ 384 pages- Nemotron 3 Nano 30B A3B (free)NVIDIAFree256K256,000≈ 384256K · 256,000 tokens · ≈ 384 pages
- Nemotron 3 Nano Omni (free)NVIDIAFree256K256,000≈ 384256K · 256,000 tokens · ≈ 384 pages
- North Mini Code (free)CohereFree256K256,000≈ 384256K · 256,000 tokens · ≈ 384 pages
- Qwen3 VL 8B InstructQwen256K256,000≈ 384256K · 256,000 tokens · ≈ 384 pages
- Qwen3 VL 8B ThinkingQwen256K256,000≈ 384256K · 256,000 tokens · ≈ 384 pages
- Qwen3.5 397B A17BQwen256K256,000≈ 384256K · 256,000 tokens · ≈ 384 pages
Relace Apply 3Relace256K256,000≈ 384256K · 256,000 tokens · ≈ 384 pages- Relace SearchRelace256K256,000≈ 384256K · 256,000 tokens · ≈ 384 pages
- Step 3.7 FlashStepFun256K256,000≈ 384256K · 256,000 tokens · ≈ 384 pages
- MiniMax M2MiniMax205K204,800≈ 307205K · 204,800 tokens · ≈ 307 pages
- MiniMax M2.1MiniMax205K204,800≈ 307205K · 204,800 tokens · ≈ 307 pages
- MiniMax M2.5MiniMax205K204,800≈ 307205K · 204,800 tokens · ≈ 307 pages
- MiniMax M2.7MiniMax205K204,800≈ 307205K · 204,800 tokens · ≈ 307 pages
- GLM 4.6Z.ai203K202,752≈ 304203K · 202,752 tokens · ≈ 304 pages
- GLM 4.7Z.ai203K202,752≈ 304203K · 202,752 tokens · ≈ 304 pages
- GLM 4.7 FlashZ.ai203K202,752≈ 304203K · 202,752 tokens · ≈ 304 pages
- GLM 5Z.ai203K202,752≈ 304203K · 202,752 tokens · ≈ 304 pages
- GLM 5.1Z.ai203K202,752≈ 304203K · 202,752 tokens · ≈ 304 pages
- GLM 5V TurboZ.ai203K202,752≈ 304203K · 202,752 tokens · ≈ 304 pages
- Anthropic Claude Haiku LatestAnthropic200K200,000≈ 300200K · 200,000 tokens · ≈ 300 pages
- Claude 3 HaikuAnthropic200K200,000≈ 300200K · 200,000 tokens · ≈ 300 pages
- Claude Haiku 4.5Anthropic200K200,000≈ 300200K · 200,000 tokens · ≈ 300 pages
- Claude Opus 4Anthropic200K200,000≈ 300200K · 200,000 tokens · ≈ 300 pages
- Claude Opus 4.1Anthropic200K200,000≈ 300200K · 200,000 tokens · ≈ 300 pages
- Claude Opus 4.5Anthropic200K200,000≈ 300200K · 200,000 tokens · ≈ 300 pages
- Free Models RouterOpenRouterFree200K200,000≈ 300200K · 200,000 tokens · ≈ 300 pages
- o1OpenAI200K200,000≈ 300200K · 200,000 tokens · ≈ 300 pages
- o1-proOpenAI200K200,000≈ 300200K · 200,000 tokens · ≈ 300 pages
- o3OpenAI200K200,000≈ 300200K · 200,000 tokens · ≈ 300 pages
- o3 Deep ResearchOpenAI200K200,000≈ 300200K · 200,000 tokens · ≈ 300 pages
- o3 MiniOpenAI200K200,000≈ 300200K · 200,000 tokens · ≈ 300 pages
- o3 Mini HighOpenAI200K200,000≈ 300200K · 200,000 tokens · ≈ 300 pages
- o3 ProOpenAI200K200,000≈ 300200K · 200,000 tokens · ≈ 300 pages
- o4 MiniOpenAI200K200,000≈ 300200K · 200,000 tokens · ≈ 300 pages
- o4 Mini Deep ResearchOpenAI200K200,000≈ 300200K · 200,000 tokens · ≈ 300 pages
- o4 Mini HighOpenAI200K200,000≈ 300200K · 200,000 tokens · ≈ 300 pages
Sonar ProPerplexity200K200,000≈ 300200K · 200,000 tokens · ≈ 300 pages- Sonar Pro SearchPerplexity200K200,000≈ 300200K · 200,000 tokens · ≈ 300 pages
- DeepSeek V3 0324DeepSeek164K163,840≈ 246164K · 163,840 tokens · ≈ 246 pages
- DeepSeek V3.1DeepSeek164K163,840≈ 246164K · 163,840 tokens · ≈ 246 pages
- DeepSeek V3.1 TerminusDeepSeek164K163,840≈ 246164K · 163,840 tokens · ≈ 246 pages
- DeepSeek V3.2 ExpDeepSeek164K163,840≈ 246164K · 163,840 tokens · ≈ 246 pages
- Llama Guard 4 12BMeta AI164K163,840≈ 246164K · 163,840 tokens · ≈ 246 pages
- R1DeepSeek164K163,840≈ 246164K · 163,840 tokens · ≈ 246 pages
- R1 0528DeepSeek164K163,840≈ 246164K · 163,840 tokens · ≈ 246 pages
- Qwen3 Coder 30B A3B InstructQwen160K160,000≈ 240160K · 160,000 tokens · ≈ 240 pages
- Qwen3 14BQwen132K131,702≈ 198132K · 131,702 tokens · ≈ 198 pages
Aion-1.0Aion Labs131K131,072≈ 197131K · 131,072 tokens · ≈ 197 pages- Aion-1.0-MiniAion Labs131K131,072≈ 197131K · 131,072 tokens · ≈ 197 pages
- Aion-2.0Aion Labs131K131,072≈ 197131K · 131,072 tokens · ≈ 197 pages
Cydonia 24B V4.1TheDrummer131K131,072≈ 197131K · 131,072 tokens · ≈ 197 pages- DeepSeek V3DeepSeek131K131,072≈ 197131K · 131,072 tokens · ≈ 197 pages
- DeepSeek V3.2DeepSeek131K131,072≈ 197131K · 131,072 tokens · ≈ 197 pages
ERNIE 4.5 VL 424B A47B Baidu131K131,072≈ 197131K · 131,072 tokens · ≈ 197 pages- Gemma 3 12BGoogle131K131,072≈ 197131K · 131,072 tokens · ≈ 197 pages
- Gemma 3 27BGoogle131K131,072≈ 197131K · 131,072 tokens · ≈ 197 pages
- Gemma 3 4BGoogle131K131,072≈ 197131K · 131,072 tokens · ≈ 197 pages
- GLM 4.5Z.ai131K131,072≈ 197131K · 131,072 tokens · ≈ 197 pages
- GLM 4.5 AirZ.ai131K131,072≈ 197131K · 131,072 tokens · ≈ 197 pages
- GLM 4.6VZ.ai131K131,072≈ 197131K · 131,072 tokens · ≈ 197 pages
- gpt-oss-120bOpenAI131K131,072≈ 197131K · 131,072 tokens · ≈ 197 pages
- gpt-oss-120b (free)OpenAIFree131K131,072≈ 197131K · 131,072 tokens · ≈ 197 pages
- gpt-oss-20bOpenAI131K131,072≈ 197131K · 131,072 tokens · ≈ 197 pages
- gpt-oss-20b (free)OpenAIFree131K131,072≈ 197131K · 131,072 tokens · ≈ 197 pages
- gpt-oss-safeguard-20bOpenAI131K131,072≈ 197131K · 131,072 tokens · ≈ 197 pages
Granite 4.1 8BIBM131K131,072≈ 197131K · 131,072 tokens · ≈ 197 pages
Hermes 3 405B InstructNous Research131K131,072≈ 197131K · 131,072 tokens · ≈ 197 pages- Hermes 3 405B Instruct (free)Nous ResearchFree131K131,072≈ 197131K · 131,072 tokens · ≈ 197 pages
- Hermes 3 70B InstructNous Research131K131,072≈ 197131K · 131,072 tokens · ≈ 197 pages
- Hermes 4 405BNous Research131K131,072≈ 197131K · 131,072 tokens · ≈ 197 pages
- Hermes 4 70BNous Research131K131,072≈ 197131K · 131,072 tokens · ≈ 197 pages
- Hunyuan A13B InstructTencent131K131,072≈ 197131K · 131,072 tokens · ≈ 197 pages
- Kimi K2 0711Moonshot AI131K131,072≈ 197131K · 131,072 tokens · ≈ 197 pages
- Llama 3.1 70B InstructMeta AI131K131,072≈ 197131K · 131,072 tokens · ≈ 197 pages
- Llama 3.1 8B InstructMeta AI131K131,072≈ 197131K · 131,072 tokens · ≈ 197 pages
Llama 3.1 Euryale 70B v2.2Sao10K131K131,072≈ 197131K · 131,072 tokens · ≈ 197 pages- Llama 3.2 11B Vision InstructMeta AI131K131,072≈ 197131K · 131,072 tokens · ≈ 197 pages
- Llama 3.2 1B InstructMeta AI131K131,072≈ 197131K · 131,072 tokens · ≈ 197 pages
- Llama 3.2 3B InstructMeta AI131K131,072≈ 197131K · 131,072 tokens · ≈ 197 pages
- Llama 3.2 3B Instruct (free)Meta AIFree131K131,072≈ 197131K · 131,072 tokens · ≈ 197 pages
- Llama 3.3 70B InstructMeta AI131K131,072≈ 197131K · 131,072 tokens · ≈ 197 pages
- Llama 3.3 70B Instruct (free)Meta AIFree131K131,072≈ 197131K · 131,072 tokens · ≈ 197 pages
- Llama 3.3 Euryale 70BSao10K131K131,072≈ 197131K · 131,072 tokens · ≈ 197 pages
- Llama 3.3 Nemotron Super 49B V1.5NVIDIA131K131,072≈ 197131K · 131,072 tokens · ≈ 197 pages
- Ministral 3 3B 2512Mistral AI131K131,072≈ 197131K · 131,072 tokens · ≈ 197 pages
- Mistral Large 2407Mistral AI131K131,072≈ 197131K · 131,072 tokens · ≈ 197 pages
- Mistral Medium 3Mistral AI131K131,072≈ 197131K · 131,072 tokens · ≈ 197 pages
- Mistral Medium 3.1Mistral AI131K131,072≈ 197131K · 131,072 tokens · ≈ 197 pages
- Mistral NemoMistral AI131K131,072≈ 197131K · 131,072 tokens · ≈ 197 pages
- Nano Banana 2 (Gemini 3.1 Flash Image Preview)Google131K131,072≈ 197131K · 131,072 tokens · ≈ 197 pages
- Nano Banana 2 (Gemini 3.1 Flash Image)Google131K131,072≈ 197131K · 131,072 tokens · ≈ 197 pages
- Qwen2.5 72B InstructQwen131K131,072≈ 197131K · 131,072 tokens · ≈ 197 pages
- Qwen2.5 7B InstructQwen131K131,072≈ 197131K · 131,072 tokens · ≈ 197 pages
- Qwen2.5 VL 72B InstructQwen131K131,072≈ 197131K · 131,072 tokens · ≈ 197 pages
- Qwen3 235B A22BQwen131K131,072≈ 197131K · 131,072 tokens · ≈ 197 pages
- Qwen3 30B A3BQwen131K131,072≈ 197131K · 131,072 tokens · ≈ 197 pages
- Qwen3 30B A3B Instruct 2507Qwen131K131,072≈ 197131K · 131,072 tokens · ≈ 197 pages
- Qwen3 30B A3B Thinking 2507Qwen131K131,072≈ 197131K · 131,072 tokens · ≈ 197 pages
- Qwen3 32BQwen131K131,072≈ 197131K · 131,072 tokens · ≈ 197 pages
- Qwen3 8BQwen131K131,072≈ 197131K · 131,072 tokens · ≈ 197 pages
- Qwen3 VL 235B A22B ThinkingQwen131K131,072≈ 197131K · 131,072 tokens · ≈ 197 pages
- Qwen3 VL 30B A3B ThinkingQwen131K131,072≈ 197131K · 131,072 tokens · ≈ 197 pages
Switchpoint RouterSwitchpoint AI131K131,072≈ 197131K · 131,072 tokens · ≈ 197 pages- Trinity MiniArcee AI131K131,072≈ 197131K · 131,072 tokens · ≈ 197 pages
- Virtuoso LargeArcee AI131K131,072≈ 197131K · 131,072 tokens · ≈ 197 pages
- Granite 4.0 MicroIBM131K131,000≈ 197131K · 131,000 tokens · ≈ 197 pages
- Body Builder (beta)OpenRouter128K128,000≈ 192128K · 128,000 tokens · ≈ 192 pages
Cogito v2.1 671BDeep Cogito128K128,000≈ 192128K · 128,000 tokens · ≈ 192 pages- Command R (08-2024)Cohere128K128,000≈ 192128K · 128,000 tokens · ≈ 192 pages
- Command R+ (08-2024)Cohere128K128,000≈ 192128K · 128,000 tokens · ≈ 192 pages
- Command R7B (12-2024)Cohere128K128,000≈ 192128K · 128,000 tokens · ≈ 192 pages
- GPT AudioOpenAI128K128,000≈ 192128K · 128,000 tokens · ≈ 192 pages
- GPT Audio MiniOpenAI128K128,000≈ 192128K · 128,000 tokens · ≈ 192 pages
- GPT-4 TurboOpenAI128K128,000≈ 192128K · 128,000 tokens · ≈ 192 pages
- GPT-4 Turbo PreviewOpenAI128K128,000≈ 192128K · 128,000 tokens · ≈ 192 pages
- GPT-4oOpenAI128K128,000≈ 192128K · 128,000 tokens · ≈ 192 pages
- GPT-4o (2024-05-13)OpenAI128K128,000≈ 192128K · 128,000 tokens · ≈ 192 pages
- GPT-4o (2024-08-06)OpenAI128K128,000≈ 192128K · 128,000 tokens · ≈ 192 pages
- GPT-4o (2024-11-20)OpenAI128K128,000≈ 192128K · 128,000 tokens · ≈ 192 pages
- GPT-4o Search PreviewOpenAI128K128,000≈ 192128K · 128,000 tokens · ≈ 192 pages
- GPT-4o-miniOpenAI128K128,000≈ 192128K · 128,000 tokens · ≈ 192 pages
- GPT-4o-mini (2024-07-18)OpenAI128K128,000≈ 192128K · 128,000 tokens · ≈ 192 pages
- GPT-4o-mini Search PreviewOpenAI128K128,000≈ 192128K · 128,000 tokens · ≈ 192 pages
- GPT-5 ChatOpenAI128K128,000≈ 192128K · 128,000 tokens · ≈ 192 pages
- GPT-5.1 ChatOpenAI128K128,000≈ 192128K · 128,000 tokens · ≈ 192 pages
- GPT-5.2 ChatOpenAI128K128,000≈ 192128K · 128,000 tokens · ≈ 192 pages
- GPT-5.3 ChatOpenAI128K128,000≈ 192128K · 128,000 tokens · ≈ 192 pages
LFM2-24B-A2BLiquid AI128K128,000≈ 192128K · 128,000 tokens · ≈ 192 pages
Mercury 2Inception128K128,000≈ 192128K · 128,000 tokens · ≈ 192 pages- Mistral LargeMistral AI128K128,000≈ 192128K · 128,000 tokens · ≈ 192 pages
- Mistral Small 3.1 24BMistral AI128K128,000≈ 192128K · 128,000 tokens · ≈ 192 pages
- Mistral Small 3.2 24BMistral AI128K128,000≈ 192128K · 128,000 tokens · ≈ 192 pages
- Nemotron 3.5 Content Safety (free)NVIDIAFree128K128,000≈ 192128K · 128,000 tokens · ≈ 192 pages
- Nemotron Nano 12B 2 VL (free)NVIDIAFree128K128,000≈ 192128K · 128,000 tokens · ≈ 192 pages
- Nemotron Nano 9B V2 (free)NVIDIAFree128K128,000≈ 192128K · 128,000 tokens · ≈ 192 pages
- Nova Micro 1.0Amazon128K128,000≈ 192128K · 128,000 tokens · ≈ 192 pages
- Qwen2.5 Coder 32B InstructQwen128K128,000≈ 192128K · 128,000 tokens · ≈ 192 pages
- R1 Distill Llama 70BDeepSeek128K128,000≈ 192128K · 128,000 tokens · ≈ 192 pages
Solar Pro 3Upstage128K128,000≈ 192128K · 128,000 tokens · ≈ 192 pages- Sonar Deep ResearchPerplexity128K128,000≈ 192128K · 128,000 tokens · ≈ 192 pages
- Sonar Reasoning ProPerplexity128K128,000≈ 192128K · 128,000 tokens · ≈ 192 pages
UI-TARS 7B ByteDance128K128,000≈ 192128K · 128,000 tokens · ≈ 192 pages- SonarPerplexity127K127,072≈ 191127K · 127,072 tokens · ≈ 191 pages
- Morph V3 FastMorph82K81,920≈ 12382K · 81,920 tokens · ≈ 123 pages
- GLM 4.5VZ.ai66K65,536≈ 9866K · 65,536 tokens · ≈ 98 pages
- MiniMax M2-herMiniMax66K65,536≈ 9866K · 65,536 tokens · ≈ 98 pages
- Mixtral 8x22B InstructMistral AI66K65,536≈ 9866K · 65,536 tokens · ≈ 98 pages
- Nano Banana 2 Lite (Gemini 3.1 Flash Lite Image)Google66K65,536≈ 9866K · 65,536 tokens · ≈ 98 pages
- Nano Banana Pro (Gemini 3 Pro Image Preview)Google66K65,536≈ 9866K · 65,536 tokens · ≈ 98 pages
- Nano Banana Pro (Gemini 3 Pro Image)Google66K65,536≈ 9866K · 65,536 tokens · ≈ 98 pages
Olmo 3 32B ThinkAi266K65,536≈ 9866K · 65,536 tokens · ≈ 98 pages
Reka Flash 3Reka66K65,536≈ 9866K · 65,536 tokens · ≈ 98 pages
WizardLM-2 8x22BMicrosoft66K65,536≈ 9866K · 65,536 tokens · ≈ 98 pages- Aion-RP 1.0 (8B)Aion Labs33K32,768≈ 4933K · 32,768 tokens · ≈ 49 pages
- Coder LargeArcee AI33K32,768≈ 4933K · 32,768 tokens · ≈ 49 pages
- Gemma 3n 4BGoogle33K32,768≈ 4933K · 32,768 tokens · ≈ 49 pages
- LFM2.5-1.2B-Instruct (free)Liquid AIFree33K32,768≈ 4933K · 32,768 tokens · ≈ 49 pages
- LFM2.5-1.2B-Thinking (free)Liquid AIFree33K32,768≈ 4933K · 32,768 tokens · ≈ 49 pages
Magnum v4 72BAnthracite33K32,768≈ 4933K · 32,768 tokens · ≈ 49 pages- Mistral Small 3Mistral AI33K32,768≈ 4933K · 32,768 tokens · ≈ 49 pages
- Nano Banana (Gemini 2.5 Flash Image)Google33K32,768≈ 4933K · 32,768 tokens · ≈ 49 pages
Perceptron Mk1Perceptron AI33K32,768≈ 4933K · 32,768 tokens · ≈ 49 pages- Rocinante 12BTheDrummer33K32,768≈ 4933K · 32,768 tokens · ≈ 49 pages
- SabaMistral AI33K32,768≈ 4933K · 32,768 tokens · ≈ 49 pages
- Skyfall 36B V2TheDrummer33K32,768≈ 4933K · 32,768 tokens · ≈ 49 pages
Uncensored (free)Cognitive ComputationsFree33K32,768≈ 4933K · 32,768 tokens · ≈ 49 pages- UnslopNemo 12BTheDrummer33K32,768≈ 4933K · 32,768 tokens · ≈ 49 pages
- Voxtral Small 24B 2507Mistral AI32K32,000≈ 4832K · 32,000 tokens · ≈ 48 pages
- GPT-3.5 TurboOpenAI16K16,385≈ 2516K · 16,385 tokens · ≈ 25 pages
- GPT-3.5 Turbo 16kOpenAI16K16,385≈ 2516K · 16,385 tokens · ≈ 25 pages
- Phi 4Microsoft16K16,384≈ 2516K · 16,384 tokens · ≈ 25 pages
- Reka EdgeReka16K16,384≈ 2516K · 16,384 tokens · ≈ 25 pages
- Llama 3.1 70B Hanami x1Sao10K16K16,000≈ 2416K · 16,000 tokens · ≈ 24 pages
- Gemma 2 27BGoogle8K8,192≈ 128K · 8,192 tokens · ≈ 12 pages
- Llama 3 8B InstructMeta AI8K8,192≈ 128K · 8,192 tokens · ≈ 12 pages
- Llama 3 8B LunarisSao10K8K8,192≈ 128K · 8,192 tokens · ≈ 12 pages
- GPT-4OpenAI8K8,191≈ 128K · 8,191 tokens · ≈ 12 pages
Inflection 3 PiInflection AI8K8,000≈ 128K · 8,000 tokens · ≈ 12 pages- Inflection 3 ProductivityInflection AI8K8,000≈ 128K · 8,000 tokens · ≈ 12 pages
Weaver (alpha)Mancer8K8,000≈ 128K · 8,000 tokens · ≈ 12 pages
ReMM SLERP 13BUndi956K6,144≈ 96K · 6,144 tokens · ≈ 9 pages
MythoMax 13BGryphe4K4,096≈ 64K · 4,096 tokens · ≈ 6 pages- GPT-3.5 Turbo (older v0613)OpenAI4K4,095≈ 64K · 4,095 tokens · ≈ 6 pages
- GPT-3.5 Turbo InstructOpenAI4K4,095≈ 64K · 4,095 tokens · ≈ 6 pages
Pages assume ~0.75 words per token and 500 words per page — an approximation; real capacity varies by tokenizer and content. Each row links to the model's page with full specs, pricing, and code samples.
Keep going
Pick a window. Bring one key.
Every model here runs through one OpenAI-compatible endpoint at its provider price plus 3%.