CAP
Long Context
League table — best at long context · 30
mean percentile across this capability's benchmarks| # | Model | League | Coverage | $/1M | tok/s |
|---|---|---|---|---|---|
| 1 | GPT-5.2 Codex | 100.0 | 1/1 | $4.81 | 144 |
| 2 | GPT-5 | 99.7 | 1/1 | $3.44 | 102 |
| 3 | GPT-5.1 | 99.4 | 1/1 | $3.44 | 114 |
| 4 | GPT-5.5 | 99.1 | 1/1 | $11.25 | 72 |
| 5 | Claude Opus 4.5 | 98.8 | 1/1 | $10 | 57 |
| 6 | GPT-5.4 | 98.5 | 1/1 | $5.63 | 161 |
| 7 | KAT-Coder-Pro V1 | 98.2 | 1/1 | $0.525 | 109 |
| 8 | MiniMax-M3 | 97.9 | 1/1 | $0.525 | 111 |
| 9 | GPT-5.6 Luna (max) | 97.6 | 1/1 | $2.25 | 234 |
| 10 | GPT-5.3 Codex | 97.3 | 1/1 | $4.81 | 99 |
| 11 | GPT-5.6 Terra (max) | 97.0 | 1/1 | $5.63 | 166 |
| 12 | GPT-5.6 Sol (max) | 96.7 | 1/1 | $11.25 | 85 |
| 13 | MiMo-V2.5-Pro | 96.4 | 1/1 | $0.544 | 55 |
| 14 | GPT-5.2 | 96.2 | 1/1 | $4.81 | 81 |
| 15 | Gemini 3.1 Pro Preview | 95.9 | 1/1 | $4.5 | 140 |
| 16 | GLM-5.2 (max) | 95.6 | 1/1 | $2.15 | 200 |
| 17 | GPT-5.6 Terra | 95.3 | 1/1 | $5.63 | 130 |
| 18 | GPT-5.6 Sol | 95.0 | 1/1 | $11.25 | 71 |
| 19 | Claude Opus 4.6 | 94.7 | 1/1 | $10 | 48 |
| 20 | Gemini 3 Pro Preview | 94.4 | 1/1 | $4.5 | — |
| 21 | Claude Sonnet 4.6 | 94.1 | 1/1 | $6 | 66 |
| 22 | Claude Sonnet 5 | 93.8 | 1/1 | $4 | 79 |
| 23 | Claude Opus 4.7 | 93.5 | 1/1 | $10 | 52 |
| 24 | Claude 4.5 Haiku | 93.2 | 1/1 | $2 | 129 |
| 25 | Claude Fable 5 | 92.9 | 1/1 | $20 | 65 |
| 26 | Qwen3.6 Max Preview | 92.6 | 1/1 | $2.92 | 53 |
| 27 | Qwen3.6 Plus | 92.3 | 1/1 | $1.13 | 53 |
| 28 | Kimi K2.6 | 92.0 | 1/1 | $1.71 | 43 |
| 29 | Muse Spark | 91.7 | 1/1 | — | — |
| 30 | GPT-5.6 Luna | 91.4 | 1/1 | $2.25 | 222 |
League = a model's mean percentile across the 1 benchmarkin this capability (best score per benchmark; partial coverage shown). Price & speed via Artificial Analysis.
Benchmarks in this capability (1)
AA Long-Context Reasoning339 scores