PostTrain BenchBrowse 296

PostTrain Bench

Can the model complete substantial programming work under this benchmark's agent setup?

Version not specifiedExact reported variant
Coding8 ranked models11 reported values2 reportsHigher is better

Top rankings

One row per model, using its best reported score across effort settings.

8
RankModelBest scoreBest reported setting
1Claude Fable 5Anthropic41.4%max effortMoonshot AI report1 value · 1 reportJul 17, 2026 · source
2Claude Opus 4.8Anthropic37.2%Reported configurationZ.ai report2 values · 2 reportsJun 16, 2026 · source
3Kimi K3Moonshot AI36.6%max effortMoonshot AI report1 value · 1 reportJul 17, 2026 · source
4GPT-5.6 SolOpenAI34.6%max effortMoonshot AI report1 value · 1 reportJul 17, 2026 · source
5GLM-5.2Z.ai34.3%max effortMoonshot AI report2 values · 2 reportsJul 17, 2026 · source
6GPT-5.5OpenAI28.4%xhigh effortMoonshot AI report2 values · 2 reportsJul 17, 2026 · source
7Gemini 3.1 ProGoogle21.6%Reported configurationZ.ai report1 value · 1 reportJun 16, 2026 · source
8GLM-5.1Z.ai20.1%Reported configurationZ.ai report1 value · 1 reportJun 16, 2026 · source

Effort curve

Every sourced cost-linked effort value for this exact version. Lines connect complete sweeps only.

0
No cost-linked effort sweep for this version.
Definition and comparison boundarypublic methodology

Can the model complete substantial programming work under this benchmark's agent setup?

Software implementation, debugging, or repository work under the published evaluation protocol. Higher is better. The value is the percentage reported in this lab's table.

Harness, tools, prompts, attempt count, and benchmark version can materially change the result. Compare within one reporting context.

Claude Code harness.

Method / source