# SWE-Bench Multilingual — AI model rankings

Like SWE-Bench, but the coding problems span many programming languages, not just one. Tests how broadly the AI can code. Higher is better.

10 tracked models have a published SWE-Bench Multilingual score. Higher is better. Scores are as published at each model's release.

## Ranking

| Rank | Model | Developer | Score | Released |
| --- | --- | --- | --- | --- |
| 1 | Claude Opus 4.8 | Anthropic | 84.4% | May 28 2026 |
| 2 | Claude Opus 4.7 | Anthropic | 80.5% | Apr 16 2026 |
| 3 | Composer 2.5 | Cursor | 79.8% | May 18 2026 |
| 4 | Grok 4.5 | SpaceXAI | 78% | Jul 8 2026 |
| 5 | GPT-5.5 | OpenAI | 77.8% | Apr 23 2026 |
| 6 | Composer 2 | Cursor | 73.7% | Mar 19 2026 |
| 7 | GLM-5 | Z.ai | 73.3% | Feb 12 2026 |
| 8 | Qwen3.5 | Qwen | 69.3% | Feb 16 2026 |
| 9 | Qwen3.6 | Qwen | 67.2% | Apr 16 2026 |
| 10 | Qwen3-Coder-Next | Qwen | 62.8% | Feb 3 2026 |


---

Canonical page: https://aireleasetracker.com/benchmark/swe-bench-multilingual
Full dataset: https://aireleasetracker.com/llms-full.txt · JSON: https://aireleasetracker.com/models.json
Source: AI Release Tracker (https://aireleasetracker.com). Benchmark scores are the figures published by the releasing lab at launch.