MMODELYST
Capabilities/Long Context
CAP

Long Context

League table — best at long context · 30
mean percentile across this capability's benchmarks
#ModelLeagueCoverage$/1Mtok/s
1GPT-5.2 Codex100.01/1$4.81
2GPT-599.71/1$3.44
3GPT-5.199.41/1$3.44
4Kimi K399.11/1$633
5GPT-5.598.91/1$11.25
6Claude Opus 4.598.61/1$10
7GPT-5.498.31/1$5.63
8KAT-Coder-Pro V198.01/1
9MiniMax-M397.71/1$0.52587
10GPT-5.3 Codex97.41/1$4.81126
11GPT-5.6 Luna (max)97.11/1$2.25171
12GPT-5.6 Terra (max)96.81/1$5.63128
13GPT-5.6 Sol (max)96.61/1$11.2574
14MiMo-V2.5-Pro96.31/1$0.54465
15GPT-5.296.01/1$4.81
16Gemini 3.1 Pro Preview95.71/1$4.5132
17GLM-5.2 (max)95.41/1$2.15157
18GPT-5.6 Terra95.11/1$5.63120
19GPT-5.6 Sol94.81/1$11.2564
20Claude Opus 4.694.51/1$10
21Gemini 3 Pro Preview94.31/1$4.5
22Claude Sonnet 4.694.01/1$6
23Claude Sonnet 593.71/1$483
24Claude Opus 4.793.41/1$10
25Claude 4.5 Haiku93.11/1$2150
26Claude Opus 592.81/1$1044
27Claude Fable 592.51/1$2058
28Qwen3.6 Max Preview92.21/1$2.92
29Qwen3.6 Plus92.01/1$1.1353
30Kimi K2.691.71/1$1.71

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)