# DeepSWE 1.1 — AI model rankings

Artificial Analysis' independent test of deep, agentic software-engineering work — the AI has to plan and carry out substantial coding tasks end to end. (Version 1.1 of the test.) Higher is better.

37 tracked models have a published DeepSWE 1.1 score. Higher is better. Scores come from published lab reports and benchmark sources; recorded sources appear beside the scores.

## Ranking

| Rank | Model | Developer | Score | Source | Released |
| --- | --- | --- | --- | --- | --- |
| 1 | Muse Wonder Full | Meta | 79% | Lab | Oct 12 2026 |
| 2 | Muse Wonder | Meta | 78.8% | Lab | Oct 12 2026 |
| 3 | Gemini 4 Argon | Google | 77.9% | Lab | Sep 30 2026 |
| 4 | DeepSeek-V4.1-Flash-1017 | DeepSeek | 77% | Lab | Oct 17 2026 |
| 5 | Kimi K3.1 Pro | Moonshot AI | 75.5% | Lab | Oct 23 2026 |
| 6 | Muse Spark 1.3 | Meta | 75.4% | Lab | Sep 2 2026 |
| 7 | GPT-6.1 Sol | OpenAI | 75.2% | Lab | Sep 29 2026 |
| 8 | DeepSeek-V4.1-Flash | DeepSeek | 74.2% | Lab | Sep 10 2026 |
| 9 | GPT-6 Astra | OpenAI | 74.1% | Lab | Sep 3 2026 |
| 10 | GPT-5.6 Sol | OpenAI | 73% | Lab | Jun 26 2026 |
| 11 | Gemini 3.8 Flash | Google | 71% | Lab | Sep 2 2026 |
| 11 | Grok 4.7 | SpaceXSI | 71% | Lab | Sep 21 2026 |
| 13 | Claude Fable 5 | Anthropic | 70% | Lab | Jun 9 2026 |
| 14 | Qwen3.8-Max-0902 | Qwen | 69.3% | Lab | Sep 2 2026 |
| 15 | Kimi K3 | Moonshot AI | 69% | Lab | Jul 16 2026 |
| 16 | Claude Opus 5 | Anthropic | 68.8% | Lab | Jul 24 2026 |
| 16 | GPT-6 Sol | OpenAI | 68.8% | Lab | Sep 22 2026 |
| 18 | GLM-5.3 | Z.ai | 66.9% | Lab | Aug 14 2026 |
| 19 | GPT-6 Luna | OpenAI | 66.6% | Lab | Sep 22 2026 |
| 20 | Grok 4.6 | SpaceXSI | 65.9% | Lab | Aug 12 2026 |
| 21 | Gemini 3.7 Flash | Google | 65.3% | Lab | Aug 13 2026 |
| 22 | GLM-5.3-Flash | Z.ai | 63.4% | Lab | Aug 26 2026 |
| 23 | Mistral Large 4 | Mistral | 61.7% | Lab | Oct 6 2026 |
| 24 | Muse Spark Open | Meta | 60% | Lab | Oct 15 2026 |
| 25 | Muse Spark 1.2 | Meta | 59.3% | Lab | Aug 5 2026 |
| 26 | Qwen3.8-Flash-Next | Qwen | 58.7% | Lab | Aug 26 2026 |
| 27 | Qwen3.8-Max | Qwen | 56.6% | Lab | Aug 3 2026 |
| 28 | Grok 4.5 | SpaceXSI | 54% | Lab | Jul 8 2026 |
| 29 | Muse Spark 1.1 | Meta | 53.3% | Lab | Jul 9 2026 |
| 30 | Gemini 3.6 Flash | Google | 49% | Lab | Jul 21 2026 |
| 31 | GLM-5.2 | Z.ai | 44% | Lab | Jun 16 2026 |
| 32 | Qwen3.8-27B | Qwen | 42.2% | Lab | Aug 14 2026 |
| 33 | Gemini 3.5 Flash | Google | 37% | Lab | May 19 2026 |
| 34 | Kimi K2.7 Code | Moonshot AI | 31% | Lab | Jun 12 2026 |
| 35 | Claude Sonnet 4.6 | Anthropic | 30% | Lab | Feb 17 2026 |
| 36 | Gemini 3.1 Pro | Google | 12% | Lab | Feb 19 2026 |
| 37 | Muse Spark | Meta | 10% | Lab | Apr 8 2026 |


---

Canonical page: https://aireleasetracker.com/benchmark/deepswe-1.1
Site index: https://aireleasetracker.com/llms.txt
Source: AI Release Tracker (https://aireleasetracker.com). Most benchmark scores come from lab launch material; gathered results identify the leaderboard that published them.