Best AI for long context
Models with a listed context window of 128K tokens or more. Ranked by composite when scored, else by context size. Full leaderboard.
| Rank | Model | Lab | Composite | Context | Out $/M |
|---|---|---|---|---|---|
| 1 | GPT 6 Astra | OpenAI | 100th | 1.1M | $50.00 |
| 2 | Claude Fable 5 | Anthropic | 99th | 1M | $50.00 |
| 3 | Claude Fable 5.1 | Anthropic | 99th | 1M | $50.00 |
| 4 | GPT-5.5 Pro | OpenAI | 98th | 1.1M | $180.00 |
| 5 | Claude Opus 5 | Anthropic | 96th | 1M | $25.00 |
| 6 | GPT-5.4 Pro | OpenAI | 95th | 1.1M | $180.00 |
| 7 | Gemini 3.8 Flash | 95th | 1M | $3.75 | |
| 8 | GPT-5.6 Sol | OpenAI | 94th | 1.1M | $10.00 |
| 9 | DeepSeek V4.1 Flash | DeepSeek | 94th | 1M | $1.2778 |
| 10 | Gemini 3.7 Flash | 94th | 1M | $3.75 | |
| 11 | GPT 5.5 | OpenAI | 93rd | 1.1M | $30.00 |
| 12 | Gemini-3-Pro | 91st | 1M | $9.60 | |
| 13 | MiniMax-M2.5 | MiniMax | 91st | 205K | $1.20 |
| 14 | GPT 5.6 Terra | OpenAI | 90th | 1.1M | $12.00 |
| 15 | Qwen3.8 Max Thinking | Alibaba | 89th | 991K | $6.00 |
| 16 | Claude Opus 4.8 | Anthropic | 89th | 1M | $25.00 |
| 17 | GPT 5.3 Codex | OpenAI | 89th | 400K | $14.00 |
| 18 | GPT 5.4 | OpenAI | 88th | 1.1M | $15.00 |
| 19 | Grok 4.6 | xAI | 87th | 500K | $6.00 |
| 20 | Gemini 3.1 Pro (Preview) | 86th | 1M | $12.00 | |
| 21 | Grok 4.5 | xAI | 85th | 500K | $6.00 |
| 22 | Gemini 3 Flash (Preview) | 84th | 1M | $3.00 | |
| 23 | Gemini 3.6 Flash | 84th | 1M | $3.75 | |
| 24 | Gemini 3.5 Flash | 84th | 1M | $9.00 | |
| 25 | Qwen3.7 Plus | Alibaba | 84th | 1M | $3.072 |
| 26 | Claude 4.7 Opus | Anthropic | 83rd | 1M | $25.00 |
| 27 | GLM-5.3 | Zhipu | 82nd | 1M | $4.40 |
| 28 | Claude Sonnet 5 Thinking | Anthropic | 80th | 1M | $10.00 |
| 29 | GPT-5.6 Luna | OpenAI | 80th | 1.1M | $0.37 |
| 30 | DeepSeek V4 Flash Vision Exp | DeepSeek | 80th | 1M | $1.32 |
| 31 | OpenAI/GPT-5.2 | OpenAI | 79th | 400K | $14.00 |
| 32 | Qwen3.7 Max | Alibaba | 79th | 1M | $9.00 |
| 33 | GPT 5 Pro | OpenAI | 78th | 400K | $120.00 |
| 34 | DeepSeek-R1 | DeepSeek | 78th | 128K | $1.70 |
| 35 | OpenAI/GPT-5 | OpenAI | 78th | 400K | $10.00 |
| 36 | GLM-5.1 | Zhipu | 77th | 200K | $4.40 |
| 37 | GPT 5.1 Codex | OpenAI | 77th | 400K | $10.00 |
| 38 | OpenAI o3 | OpenAI | 76th | 200K | $8.00 |
| 39 | GPT-5.1 (2025-11-13) | OpenAI | 75th | 400K | $10.00 |
| 40 | Claude 4.6 Opus Thinking | Anthropic | 74th | 1M | $25.00 |
| 41 | GPT 5.2 Codex | OpenAI | 73rd | 400K | $14.00 |
| 42 | Qwen3.6 35B-A3B | Alibaba | 73rd | 262K | $2.052 |
| 43 | Kimi K2 | Kimi | 72nd | 128K | free |
| 44 | Kimi K2.6 | Kimi | 72nd | 262K | $4.00 |
| 45 | Moonshotai/Kimi-K2.5 | Alibaba | 71st | 256K | $1.90 |
| 46 | Qwen3.5 397B-A17B | Alibaba | 70th | 262K | $2.65 |
| 47 | DeepSeek V3.2 | DeepSeek | 68th | 128K | $0.326 |
| 48 | Minimax/Minimax-M2 | MiniMax | 68th | 200K | $1.53 |
| 49 | GLM-5 | Alibaba | 66th | 205K | $3.2647 |
| 50 | GLM-5.2 | Alibaba | 66th | 1M | $4.40 |
| 51 | Qwen3.5 Plus Thinking | Alibaba | 65th | 984K | $2.40 |
| 52 | Kimi K2.7 Code | Kimi | 63rd | 262K | $4.389 |
| 53 | Qwen3.5 35B A3B | Alibaba | 63rd | 260K | $1.80 |
| 54 | GLM-5.3-Flash | Zhipu | 61st | 1M | $0.025 |
| 55 | Inkling Thinking | Databricks | 61st | 1M | $4.05 |
| 56 | OpenAI o1 | OpenAI | 59th | 200K | $60.00 |
| 57 | Qwen3 Max Preview | Alibaba | 58th | 256K | $6.001 |
| 58 | Gemini 2.5 Pro | 57th | 1M | $10.00 | |
| 59 | Gemini 3.1 Flash Lite | 56th | 1M | $1.50 | |
| 60 | Claude 4.5 Opus | Anthropic | 55th | 200K | $25.00 |
| 61 | OpenAI o4-mini | OpenAI | 55th | 200K | $4.40 |
| 62 | GPT 5 Mini | OpenAI | 54th | 400K | $2.00 |
| 63 | Claude Sonnet 4.5 | Anthropic | 54th | 200K | $15.00 |
| 64 | Z-Ai/GLM 4.7 | Zhipu | 51st | 200K | $0.80 |
| 65 | GPT 5.4 Mini | OpenAI | 50th | 400K | $4.50 |
| 66 | Qwen3.5 9B | Alibaba | 48th | 256K | $0.15 |
| 67 | GPT OSS 120B | OpenAI | 47th | 131K | $0.798 |
| 68 | GPT 5.4 Nano | OpenAI | 46th | 400K | $1.25 |
| 69 | Qwen3.6 Flash | Alibaba | 46th | 992K | $1.16 |
| 70 | Qwen3.5 122B A10B Thinking | Alibaba | 45th | 131K | $3.496 |
| 71 | Claude 4.1 Opus | Anthropic | 43rd | 200K | $75.00 |
| 72 | GLM 4.7 Flash Thinking | Zhipu | 43rd | 200K | $0.40 |
| 73 | OpenAI o3-mini | OpenAI | 42nd | 200K | $4.40 |
| 74 | GPT 4.1 | OpenAI | 41st | 1M | $8.00 |
| 75 | Gemini 2.5 Flash | 40th | 1M | $2.50 | |
| 76 | GPT 5 Nano | OpenAI | 40th | 400K | $0.40 |
| 77 | Gemini 3.5 Flash Lite | 39th | 1M | $2.50 | |
| 78 | Z-AI/GLM 4.6 | Zhipu | 38th | 200K | $1.40 |
| 79 | DeepSeek-V3 | DeepSeek | 35th | 128K | $0.77 |
| 80 | DeepSeek V4 Pro | DeepSeek | 35th | 1M | $0.87 |
| 81 | GLM 4.5 | Zhipu | 34th | 131K | $1.30 |
| 82 | GPT 4.1 Mini | OpenAI | 31st | 1M | $1.60 |
| 83 | GPT 4.1 Nano | OpenAI | 22nd | 1M | $0.40 |
| 84 | GPT-4o (2024-08-06) | OpenAI | 21st | 128K | $10.00 |
| 85 | Mistral Large 2411 | Mistral | 15th | 128K | $6.001 |
| 86 | GPT-4o mini | OpenAI | 15th | 128K | $0.60 |
| 87 | Grok Build 0.1 | xAI | 12th | 256K | $2.00 |
| 88 | Grok 4.3 | xAI | 0th | 1M | $2.50 |
| 89 | X-Ai/Grok 4.1 Fast Non Reasoning | xAI | — | 2M | $0.50 |
| 90 | Gemini 2.5 Flash Lite | — | 1M | $0.40 | |
| 91 | Kimi K3 | Alibaba | — | 1M | $15.00 |
| 92 | MiMo-V2.5 | — | — | 1M | $2.00 |
| 93 | MiMo-V2.5-Pro | — | — | 1M | free |
| 94 | Claude 4 Sonnet Thinking | Anthropic | — | 1M | $15.00 |
| 95 | Claude Sonnet 4.6 | Anthropic | — | 1M | $15.00 |
| 96 | DeepSeek V4 Flash | DeepSeek | — | 1M | $0.196 |
| 97 | Qwen 3.6 Plus | Alibaba | — | 992K | $1.95 |
| 98 | Qwen3.8 Flash | Alibaba | — | 992K | $0.42 |
| 99 | MiniMax M3 Thinking | MiniMax | — | 512K | $1.20 |
| 100 | GPT 5.1 Codex Max | OpenAI | — | 400K | $20.00 |
| 101 | GPT 5.1 Codex Mini | OpenAI | — | 400K | $2.00 |
| 102 | Gemma 4 26B A4B | — | 262K | $0.38 | |
| 103 | Gemma 4 31B IT (free) | — | 262K | free | |
| 104 | Qwen3 235B A22B Instruct 2507 | Alibaba | — | 262K | $3.078 |
| 105 | Qwen3 235B A22B Thinking 2507 | Alibaba | — | 262K | $0.60 |
| 106 | Qwen3 Coder Next | Alibaba | — | 262K | $1.50 |
| 107 | Qwen3.8 27B | Alibaba | — | 262K | $0.70 |
| 108 | Step 3.7 Flash | — | — | 262K | $1.14 |
| 109 | Tencent Hy3 | — | — | 262K | $0.26 |
| 110 | Qwen3 Coder 480B A35B Instruct | Alibaba | — | 262K | $1.00 |
| 111 | Qwen3.5 27B | Alibaba | — | 260K | $2.16 |
| 112 | Qwen3.6 27B Thinking | Alibaba | — | 260K | $2.24 |
| 113 | Codestral 2508 | Mistral | — | 256K | $0.90 |
| 114 | Kimi K2 Thinking | Kimi | — | 256K | $2.50 |
| 115 | MiniMax M2.7 | MiniMax | — | 205K | $1.26 |
| 116 | MiniMax-M2.7-highspeed | MiniMax | — | 205K | $2.40 |
| 117 | Minimax/Minimax-M2.1 | MiniMax | — | 205K | $1.32 |
| 118 | GLM 5 Turbo | Zhipu | — | 203K | $4.00 |
| 119 | GLM 5V Turbo Thinking | Zhipu | — | 203K | $4.00 |
| 120 | Claude 3.7 Sonnet | Anthropic | — | 200K | $15.00 |
| 121 | Claude 4 Opus Thinking | Anthropic | — | 200K | $75.00 |
| 122 | Claude Haiku 4.5 Thinking | Anthropic | — | 200K | $5.00 |
| 123 | DeepSeek V3.2 Exp | DeepSeek | — | 164K | $0.42 |
| 124 | Llama 3.1 8b Instruct | Meta | — | 131K | $0.085 |
| 125 | Llama 3.2 3b Instruct | Meta | — | 131K | $0.0493 |
| 126 | Llama 4 Maverick 17B 128E Instruct FP8 | Meta | — | 131K | $1.484 |
| 127 | Qwen3 Next 80B A3B Instruct | Alibaba | — | 131K | $0.65 |
| 128 | Qwen3 Next 80B A3B Thinking | Alibaba | — | 131K | $0.65 |
| 129 | Qwen3 VL 235B A22B Instruct | Alibaba | — | 131K | $1.20 |
| 130 | Qwen3 VL 235B A22B Thinking | Alibaba | — | 131K | $6.00 |
| 131 | GLM 4.5 Air | Zhipu | — | 131K | $0.80 |
| 132 | Cerebras-Llama-4-Scout-17B-16E-Instruct | Meta | — | 128K | free |
| 133 | DeepSeek-V3.1 | DeepSeek | — | 128K | $0.70 |
| 134 | GPT-4 Turbo | OpenAI | — | 128K | $30.00 |
| 135 | gpt-oss-20b | OpenAI | — | 128K | $0.15 |
| 136 | Mistral Small 3.2 (Mistral AI) | Mistral | — | 128K | $0.30 |
| 137 | Qwen 3 235B A22B | Alibaba | — | 128K | $0.50 |
| 138 | Qwen3 Coder Flash | Alibaba | — | 128K | $1.50 |
| 139 | Qwen3 Coder Plus | Alibaba | — | 128K | $5.00 |
| 140 | Qwen3-Coder 30B-A3B Instruct | Alibaba | — | 128K | $1.083 |
Common questions
- What is the best ai for long context right now?
- Models with a listed context window of 128K tokens or more. Ranked by composite when scored, else by context size.
- How is this ranking computed?
- The set is a spec filter on the catalog. Scored models rank by the overall composite. Unscored models follow by the spec itself.