EnvSimBench - Simple: leaderboard

The Simple tier of EnvSimBench (2026): 50 tool calls whose correct outcome changes 1 or 2 changed state fields of the environment state. The model sees the state, the action and its code and must predict the resulting state; the score is Config Match, the percent of samples reproduced exactly, with non-thinking mode throughout; higher is better.

Metric: Config Match (%): share of the 50 Simple state-changing samples (1 or 2 changed state fields) whose predicted state changes, applied to the before-state, reproduce the executor after-state exactly; MDP prompt with the state, the action and its code, non-thinking mode; higher is better. Source: arxiv.org. Saturation forecast: Around 2030. 10 models tracked. Updated 7 Oct 2026.

Top models

#ModelSettingScoreOverall rank
1MiniMax-M2.750#539
2Gemini 3.1 Pro (Preview)48#99
3Qwen 3.5 397B A17BNon-reasoning46#264
4GLM-5Non-reasoning44#269
5Qwen 3.5 9BNon-reasoning42#828
6Qwen 3.5 4BNon-reasoning42#1241
7Claude Sonnet 4.640#173
8GPT-5.4Non-reasoning40#135
9Qwen 3.5 27BNon-reasoning26#375
10DeepSeek V3.2Non-reasoning22#419

No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=envsimbench-simple · How It Works · Data refreshed daily, snapshot 2026-11-21.