Coding benchmark

# SWE-bench Multilingual leaderboard

> SWE-bench Multilingual results for 13 AI models, led by Gemini 3 Flash Preview at 72.7%. What the benchmark measures, who runs it, and a source for every score.
- Canonical page: https://noometry.com/benchmarks/swe-bench-multilingual
- Last updated: 2026-10-10
- Title: SWE-bench Multilingual Leaderboard (October 2026): Scores by Model

As of October 2026, Gemini 3 Flash Preview has the highest published SWE-bench Multilingual score on Noometry at 72.7%, out of 13 models with results.

Last verified October 10, 2026

## About SWE-bench Multilingual

Real GitHub issues from repositories in nine programming languages other than Python, solved in the same bash-only agent.

- **Category:** [Coding](https://noometry.com/best/coding)
- **Introduced:** 2025
- **Size:** 300 tasks
- **Format:** Repository patch
- **Unit:** Percent (random guessing ≈ 0%)
- **Official site:** [www.swebench.com](https://www.swebench.com/multilingual.html)

## Top 13 models

Top models on SWE-bench Multilingual

1.  Gemini 3 Flash Preview 72.7%
2.  Claude Opus 4.6 72%
3.  Claude Opus 4.5 70.7%
4.  GLM-5 69.7%
5.  Gemini 3 Pro 68.7%
6.  MiniMax-M2.5 68.3%
7.  Kimi K2.5 67.3%
8.  Claude Sonnet 4.5 67%
9.  GPT-5.2 66.7%
10.  GPT-5.2 Codex 66.3%
11.  Claude Haiku 4.5 64.7%
12.  DeepSeek-V3.2-Exp 59%
13.  GPT-5 Mini 39.7%
14.  304050607080

Sponsored placements are available on pages like this one. [Advertise on Noometry](https://noometry.com/advertise)

## All results

SWE-bench Multilingual results by model
| # | Model | Provider | Score | Setting | Source | Date |
| --- | --- | --- | --- | --- | --- | --- |
| 1 | [Gemini 3 Flash Preview](https://noometry.com/models/gemini-3-flash-preview) |  [![](/logos/google.svg) Google](https://noometry.com/providers/google) | 72.7% |  | [SWE-bench](https://www.swebench.com/) | 2026-02-13 |
| 2 | [Claude Opus 4.6](https://noometry.com/models/claude-opus-4-6) | [Anthropic](https://noometry.com/providers/anthropic) | 72% |  | [SWE-bench](https://www.swebench.com/) | 2026-02-13 |
| 3 | [Claude Opus 4.5](https://noometry.com/models/claude-opus-4-5) | [Anthropic](https://noometry.com/providers/anthropic) | 70.7% |  | [SWE-bench](https://www.swebench.com/) | 2026-02-13 |
| 4 | [GLM-5](https://noometry.com/models/glm-5) | [Z.ai (Zhipu)](https://noometry.com/providers/zai) | 69.7% |  | [SWE-bench](https://www.swebench.com/) | 2026-02-13 |
| 5 | [Gemini 3 Pro](https://noometry.com/models/gemini-3-pro) |  [![](/logos/google.svg) Google](https://noometry.com/providers/google) | 68.7% |  | [SWE-bench](https://www.swebench.com/) | 2026-02-13 |
| 6 | [MiniMax-M2.5](https://noometry.com/models/minimax-m2-5) |  [![](/logos/minimax.svg) MiniMax](https://noometry.com/providers/minimax) | 68.3% |  | [SWE-bench](https://www.swebench.com/) | 2026-02-16 |
| 7 | [Kimi K2.5](https://noometry.com/models/kimi-k2-5) | [Moonshot AI](https://noometry.com/providers/moonshot) | 67.3% |  | [SWE-bench](https://www.swebench.com/) | 2026-02-13 |
| 8 | [Claude Sonnet 4.5](https://noometry.com/models/claude-sonnet-4-5) | [Anthropic](https://noometry.com/providers/anthropic) | 67% |  | [SWE-bench](https://www.swebench.com/) | 2026-02-13 |
| 9 | [GPT-5.2](https://noometry.com/models/gpt-5-2) | [OpenAI](https://noometry.com/providers/openai) | 66.7% | high | [SWE-bench](https://www.swebench.com/) | 2026-02-13 |
| 10 | [GPT-5.2 Codex](https://noometry.com/models/gpt-5-2-codex) | [OpenAI](https://noometry.com/providers/openai) | 66.3% |  | [SWE-bench](https://www.swebench.com/) | 2026-02-20 |
| 11 | [Claude Haiku 4.5](https://noometry.com/models/claude-haiku-4-5) | [Anthropic](https://noometry.com/providers/anthropic) | 64.7% |  | [SWE-bench](https://www.swebench.com/) | 2026-02-13 |
| 12 | [DeepSeek-V3.2-Exp](https://noometry.com/models/deepseek-v3-2-exp) |  [![](/logos/deepseek.svg) DeepSeek](https://noometry.com/providers/deepseek) | 59% |  | [SWE-bench](https://www.swebench.com/) | 2026-02-13 |
| 13 | [GPT-5 Mini](https://noometry.com/models/gpt-5-mini) | [OpenAI](https://noometry.com/providers/openai) | 39.7% |  | [SWE-bench](https://www.swebench.com/) | 2026-02-13 |

## Compare the leaders

-   [Gemini 3 Flash Preview vs Claude Opus 4.6](https://noometry.com/compare/claude-opus-4-6-vs-gemini-3-flash-preview)
-   [Gemini 3 Flash Preview vs Claude Opus 4.5](https://noometry.com/compare/claude-opus-4-5-vs-gemini-3-flash-preview)
-   [Gemini 3 Flash Preview vs GLM-5](https://noometry.com/compare/gemini-3-flash-preview-vs-glm-5)
-   [Gemini 3 Flash Preview vs Gemini 3 Pro](https://noometry.com/compare/gemini-3-flash-preview-vs-gemini-3-pro)
-   [Claude Opus 4.6 vs Claude Opus 4.5](https://noometry.com/compare/claude-opus-4-5-vs-claude-opus-4-6)
-   [Claude Opus 4.6 vs GLM-5](https://noometry.com/compare/claude-opus-4-6-vs-glm-5)

## Other coding benchmarks

-   [SWE-bench Verified](https://noometry.com/benchmarks/swe-bench-verified)
-   [DeepSWE](https://noometry.com/benchmarks/deepswe)
-   [FrontierCode](https://noometry.com/benchmarks/frontiercode)
-   [SWE-bench Verified (bash only)](https://noometry.com/benchmarks/swe-bench-bash-only)
-   [Aider Polyglot](https://noometry.com/benchmarks/aider-polyglot)
-   [LMArena WebDev](https://noometry.com/benchmarks/arena-webdev)
-   [CursorBench](https://noometry.com/benchmarks/cursorbench)
-   [FrontierSWE](https://noometry.com/benchmarks/frontierswe)
-   [SciCode](https://noometry.com/benchmarks/scicode)
-   [GSO](https://noometry.com/benchmarks/gso-bench)
-   [WeirdML](https://noometry.com/benchmarks/weirdml)
-   [LMArena Coding](https://noometry.com/benchmarks/arena-coding)

## Frequently asked questions

### What does SWE-bench Multilingual measure?

Real GitHub issues from repositories in nine programming languages other than Python, solved in the same bash-only agent.

### Which model has the highest SWE-bench Multilingual score?

As of October 2026, Gemini 3 Flash Preview has the highest published SWE-bench Multilingual score on Noometry at 72.7%, out of 13 models with results.

### What is the best open-weight model on SWE-bench Multilingual?

GLM-5 has the highest SWE-bench Multilingual accuracy among open-weight models at 69.7%, ranking 4 of 13 overall.

### Cite this page

Noometry. (2026). SWE-bench Multilingual leaderboard. Retrieved October 10, 2026, from https://noometry.com/benchmarks/swe-bench-multilingual

Quote Noometry with a link back to this page. It is also available in [Markdown](https://noometry.com/md/benchmarks/swe-bench-multilingual.md).