# PostTrainBench v1 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://modelscale.dev/leaderboards/benchmarks/coding-post-train-bench1

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

## Benchmark record

| Field | Value |
| --- | --- |
| Category | Coding |
| Measure | Weighted aggregate across four base models and seven benchmarks |
| Tasks | Post-training four base models on seven benchmarks |
| Difficulty | Autonomous ML engineering |
| Published by | [PostTrainBench v1 leaderboard (archived)](https://posttrainbench.com/?version=v1) |

The first seven-benchmark PostTrainBench board, kept by the source as an archived version.

## PostTrainBench v1 ranking

13 models with a published PostTrainBench v1 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 seven benchmarks | Evidence |
| --- | --- | --- | --- | --- |
| 1 | [GLM-5.2](https://modelscale.dev/models/glm-5-2) | Z.AI | 34.3 | Observed 2026-11-21 · source benchlm:models |
| 2 | [Claude Opus 4.8](https://modelscale.dev/models/claude-opus-4-8) | Anthropic | 34.1 | Observed 2026-11-21 · source benchlm:models |
| 3 | [Claude Opus 4.7](https://modelscale.dev/models/claude-opus-4-7) | Anthropic | 28.6 | Observed 2026-11-21 · source benchlm:models |
| 4 | [GPT-5.5](https://modelscale.dev/models/gpt-5-5) | OpenAI | 25.0 | Observed 2026-11-21 · source benchlm:models |
| 5 | [Claude Opus 4.6](https://modelscale.dev/models/claude-opus-4-6) | Anthropic | 23.2 | Observed 2026-11-21 · source benchlm:models |
| 6 | [Gemini 3.1 Pro](https://modelscale.dev/models/gemini-3-1-pro) | Google | 21.6 | Observed 2026-11-21 · source benchlm:models |
| 7 | [GPT-5.2](https://modelscale.dev/models/gpt-5-2) | OpenAI | 21.4 | Observed 2026-11-21 · source benchlm:models |
| 8 | [GPT-5.4](https://modelscale.dev/models/gpt-5-4) | OpenAI | 20.2 | Observed 2026-11-21 · source benchlm:models |
| 9 | [GPT-5.1-Codex-Max](https://modelscale.dev/models/gpt-5-1-codex-max) | OpenAI | 19.7 | Observed 2026-11-21 · source benchlm:models |
| 10 | [Gemini 3 Pro](https://modelscale.dev/models/gemini-3-pro) | Google | 18.1 | Observed 2026-11-21 · source benchlm:models |
| 11 | [GPT-5.3 Codex](https://modelscale.dev/models/gpt-5-3-codex) | OpenAI | 17.8 | Observed 2026-11-21 · source benchlm:models |
| 12 | [GPT-5.2-Codex](https://modelscale.dev/models/gpt-5-2-codex) | OpenAI | 17.2 | Observed 2026-11-21 · source benchlm:models |
| 13 | [Claude Opus 4.5](https://modelscale.dev/models/claude-opus-4-5) | Anthropic | 17.1 | 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.
