# SWE-Rebench 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-swe-rebench

Every model the catalog carries a published SWE-Rebench value for, ranked by that value.

## Benchmark record

| Field | Value |
| --- | --- |
| Category | Coding |
| Measure | Code patch generation |
| Tasks | Fresh GitHub issues (rolling window) |
| Difficulty | Professional software engineering |
| Published by | [SWE-Rebench: Contamination-Free Evaluation of Software Engineering Agents](https://swe-rebench.com/) |

A continuously updated software engineering benchmark by Nebius using fresh GitHub issues to avoid contamination. Models are evaluated 5 times per problem under a fixed ReAct scaffolding; the Resolved Rate (best pass@1) is reported.

## SWE-Rebench ranking

13 models with a published SWE-Rebench 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 | Code patch generation | Evidence |
| --- | --- | --- | --- | --- |
| 1 | [Claude Opus 4.6](https://modelchoice.dev/models/claude-opus-4-6) | Anthropic | 65.3 | Observed 2026-11-21 · source benchlm:models |
| 2 | [GLM-5](https://modelchoice.dev/models/glm-5) | Z.AI | 62.8 | Observed 2026-11-21 · source benchlm:models |
| 3 | [GLM-5.1](https://modelchoice.dev/models/glm-5-1) | Z.AI | 62.7 | Observed 2026-11-21 · source benchlm:models |
| 4 | [DeepSeek V3.2](https://modelchoice.dev/models/deepseek-v3-2) | DeepSeek | 60.9 | Observed 2026-11-21 · source benchlm:models |
| 5 | [Claude Sonnet 4.6](https://modelchoice.dev/models/claude-sonnet-4-6) | Anthropic | 60.7 | Observed 2026-11-21 · source benchlm:models |
| 6 | [Qwen3.5-27B](https://modelchoice.dev/models/qwen3-5-27b) | Alibaba | 58.9 | Observed 2026-11-21 · source benchlm:models |
| 7 | [GLM-4.7](https://modelchoice.dev/models/glm-4-7) | Z.AI | 58.7 | Observed 2026-11-21 · source benchlm:models |
| 8 | [Kimi K2.5](https://modelchoice.dev/models/kimi-k2-5) | Moonshot AI | 58.5 | Observed 2026-11-21 · source benchlm:models |
| 9 | [GPT-5.3 Codex](https://modelchoice.dev/models/gpt-5-3-codex) | OpenAI | 58.2 | Observed 2026-11-21 · source benchlm:models |
| 10 | [Composer 2](https://modelchoice.dev/models/composer-2) | Cursor | 58 | Observed 2026-11-21 · source benchlm:models |
| 11 | [Qwen3.5-35B-A3B](https://modelchoice.dev/models/qwen3-5-35b-a3b) | Alibaba | 53.7 | Observed 2026-11-21 · source benchlm:models |
| 12 | [MiniMax M2.7](https://modelchoice.dev/models/minimax-m2-7) | MiniMax | 51.9 | Observed 2026-11-21 · source benchlm:models |
| 13 | [Gemma 4 31B](https://modelchoice.dev/models/gemma-4-31b) | Google | 41.6 | 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.
