SWE-Together: leaderboard
Interactive coding-agent evaluation over 109 real multi-turn software tasks (bugfix, feature, refactor) drawn from 27 open-source repositories across TypeScript, Go, Python and Rust. Agents work through multi-turn user intents in a harness (OpenCode, mini-swe-agent); each attempt is scored by a rubric jury on goal completion, averaged across tasks and replicates.
Metric: Jury Score (%). Source: togetherbench.com. Saturation forecast: Around January 2027. 20 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Opus 5.5 | 85.18 | #3 |
| 2 | Claude Fable 5 | 85.16 | #16 |
| 3 | Grok 4.7 | 84.45 | #25 |
| 4 | Claude Fable 5.1 | 82.72 | #9 |
| 5 | Gemini 3.8 Flash | 82.17 | #21 |
| 6 | Grok 4.6 | 81.92 | #26 |
| 7 | Claude Opus 4.7 | 81.28 | #49 |
| 8 | GPT-6 Sol | 80.96 | #14 |
| 9 | Claude Opus 4.6 | 80.9 | #63 |
| 10 | Claude Opus 5 | 80.32 | #12 |
| 11 | GPT-6 Astra | 80.09 | #6 |
| 12 | GPT-6 Luna | 79.96 | #56 |
| 13 | Claude Opus 4.8 | 79.71 | #37 |
| 14 | Muse Spark 1.3 | 79.54 | #24 |
| 15 | GPT-5.6 Sol | 79.05 | #19 |
No result here: #7 GPT-6.1 Sol, #8 Gemini 4 Argon, #10 Muse Wonder Full.
Interactive version: theaggregate.ai/benchmark?slug=swe-together · How It Works · Data refreshed daily, snapshot 2026-10-20.