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$4.81144
2GPT-599.71/1$3.44102
3GPT-5.199.41/1$3.44114
4GPT-5.599.11/1$11.2572
5Claude Opus 4.598.81/1$1057
6GPT-5.498.51/1$5.63161
7KAT-Coder-Pro V198.21/1$0.525109
8MiniMax-M397.91/1$0.525111
9GPT-5.6 Luna (max)97.61/1$2.25234
10GPT-5.3 Codex97.31/1$4.8199
11GPT-5.6 Terra (max)97.01/1$5.63166
12GPT-5.6 Sol (max)96.71/1$11.2585
13MiMo-V2.5-Pro96.41/1$0.54455
14GPT-5.296.21/1$4.8181
15Gemini 3.1 Pro Preview95.91/1$4.5140
16GLM-5.2 (max)95.61/1$2.15200
17GPT-5.6 Terra95.31/1$5.63130
18GPT-5.6 Sol95.01/1$11.2571
19Claude Opus 4.694.71/1$1048
20Gemini 3 Pro Preview94.41/1$4.5
21Claude Sonnet 4.694.11/1$666
22Claude Sonnet 593.81/1$479
23Claude Opus 4.793.51/1$1052
24Claude 4.5 Haiku93.21/1$2129
25Claude Fable 592.91/1$2065
26Qwen3.6 Max Preview92.61/1$2.9253
27Qwen3.6 Plus92.31/1$1.1353
28Kimi K2.692.01/1$1.7143
29Muse Spark91.71/1
30GPT-5.6 Luna91.41/1$2.25222

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)