# PostTrainBench leaderboard

Coding benchmark

> Every figure below is reproduced as its upstream source published it: nothing is modelled, estimated, interpolated or converted. `Unavailable` means no source published the value — it is never a zero. Each value carries its evidence state and the date it was observed.

Page: https://modelchoice.dev/leaderboards/benchmarks/coding-post-train-bench12

PostTrainBench v1.2. Every model the catalog carries a published PostTrainBench value for, ranked by that value.

## Benchmark record

| Field | Value |
| --- | --- |
| Category | Coding |
| Measure | Weighted aggregate across four base models and six benchmarks |
| Tasks | Post-training four base models on six benchmarks |
| Difficulty | Autonomous ML engineering |
| Published by | [PostTrainBench](https://posttrainbench.com/) |

An AI agent gets ten hours and one H100 GPU to post-train a small base model; the score is how well the trained models do on six benchmarks.

## PostTrainBench ranking

23 models with a published PostTrainBench value, ordered by that value, highest first. Models the source has not scored on this benchmark are not listed — they are unmeasured, not last.

| Rank | Model | Provider | Weighted aggregate across four base models and six benchmarks | Evidence |
| --- | --- | --- | --- | --- |
| 1 | [Claude Opus 5.5](https://modelchoice.dev/models/claude-opus-5-5) | Anthropic | 43.8 | Observed 2026-11-21 · source benchlm:models |
| 2 | [GPT-6.2 Astra](https://modelchoice.dev/models/gpt-6-2-astra) | OpenAI | 43.5 | Observed 2026-11-21 · source benchlm:models |
| 3 | [GPT-6 Astra](https://modelchoice.dev/models/gpt-6-astra) | OpenAI | 41.9 | Observed 2026-11-21 · source benchlm:models |
| 4 | [Gemini 4 Carbon](https://modelchoice.dev/models/gemini-4-carbon) | Google | 41.0 | Observed 2026-11-21 · source benchlm:models |
| 5 | [Gemini 4 Argon](https://modelchoice.dev/models/gemini-4-argon) | Google | 40.4 | Observed 2026-11-21 · source benchlm:models |
| 6 | [GPT-6.1 Sol](https://modelchoice.dev/models/gpt-6-1-sol) | OpenAI | 38.8 | Observed 2026-11-21 · source benchlm:models |
| 7 | [GLM-5.3-Flash](https://modelchoice.dev/models/glm-5-3-flash) | Z.AI | 36.8 | Observed 2026-11-21 · source benchlm:models |
| 8 | [Qwen4 Max](https://modelchoice.dev/models/qwen4-max) | Alibaba | 36.7 | Observed 2026-11-21 · source benchlm:models |
| 9 | [Kimi K3.1 Pro](https://modelchoice.dev/models/kimi-3-1-pro) | Moonshot AI | 36.5 | Observed 2026-11-21 · source benchlm:models |
| 10 | [Claude Opus 5](https://modelchoice.dev/models/claude-opus-5) | Anthropic | 35.7 | Observed 2026-11-21 · source benchlm:models |
| 11 | [Hy4](https://modelchoice.dev/models/hy4) | Tencent | 35.2 | Observed 2026-11-21 · source benchlm:models |
| 12 | [GLM-5.3](https://modelchoice.dev/models/glm-5-3) | Z.AI | 34.5 | Observed 2026-11-21 · source benchlm:models |
| 13 | [Inkling 2 Small](https://modelchoice.dev/models/inkling-2-small) | Thinking Machines Lab | 31.9 | Observed 2026-11-21 · source benchlm:models |
| 14 | [Kimi K3](https://modelchoice.dev/models/kimi-3) | Moonshot AI | 31.5 | Observed 2026-11-21 · source benchlm:models |
| 15 | [GPT-5.6 Sol](https://modelchoice.dev/models/gpt-5-6-sol) | OpenAI | 31.0 | Observed 2026-11-21 · source benchlm:models |
| 16 | [Claude Opus 4.8](https://modelchoice.dev/models/claude-opus-4-8) | Anthropic | 31.0 | Observed 2026-11-21 · source benchlm:models |
| 17 | [GLM-5.2](https://modelchoice.dev/models/glm-5-2) | Z.AI | 30.4 | Observed 2026-11-21 · source benchlm:models |
| 18 | [Claude Opus 4.7](https://modelchoice.dev/models/claude-opus-4-7) | Anthropic | 24.3 | Observed 2026-11-21 · source benchlm:models |
| 19 | [Muse Wonder](https://modelchoice.dev/models/muse-wonder) | Meta | 24.2 | Observed 2026-11-21 · source benchlm:models |
| 20 | [GPT-5.5](https://modelchoice.dev/models/gpt-5-5) | OpenAI | 23.9 | Observed 2026-11-21 · source benchlm:models |
| 21 | [Grok 4.5](https://modelchoice.dev/models/grok-4-5) | xAI | 22.6 | Observed 2026-11-21 · source benchlm:models |
| 22 | [Gemini 3.1 Pro](https://modelchoice.dev/models/gemini-3-1-pro) | Google | 18.1 | Observed 2026-11-21 · source benchlm:models |
| 23 | [GPT-5.4](https://modelchoice.dev/models/gpt-5-4) | OpenAI | 17.8 | Observed 2026-11-21 · source benchlm:models |

Rows are ordered by the value the source published, highest first. The source does not state whether a higher value is the better result, so this document does not either: for a benchmark that measures a rate of failure, read the table from the bottom.
