Atlas

Benchmarks

← All benchmarks

PostTrainBench Average

Indexes · 2026-03-09

PostTrainBench evaluates model post-training methods across its task suite. This row reports the benchmark's weighted-average percentage score.

Top models (higher is better)

ModelScore
Kimi K336.6
GLM-5.234.3
Opus 4.834.1
Opus 4.728.6
GPT-5.527.2
Grok 4.523.4
Gemini 3.1 Pro Preview22.0
GPT-5.419.0
Loading Atlas data…