Benchmarks · reasoning
RWE-Bench
Uses average pass@1 across real-world engineering tasks, not pass@5; publication/export dates do not determine model release dates. Score: average pass@1 across engineering tasks. Collected by Jake Boggs' benchmark dashboard.
11 models · 11 results · data as of 2026-11-06 · xplore.ai ↗
Leaderboard
GPT-6 Astra max
All results
11 rows| # | Model | Effort | RWE-Bench |
|---|---|---|---|
| 1 | GPT-6 Astra | max | 37.0% |
| 2 | Opus 5.5 | max | 29.0% |
| 3 | GPT-6.1 Sol | max | 25.0% |
| 4 | Gemini 4 Argon | max | 12.0% |
| 5 | Fable 5.1 | max | 7.0% |
| 6 | Kimi K3.1 | max | 1.0% |
| 7 | Muse Wonder (Max) | max | 0.0% |
| 8 | DeepSeek V4.1 Flash | max | 0.0% |
| 9 | GLM 5.3 | max | 0.0% |
| 10 | Grok 4.7 | max | 0.0% |
| 11 | Qwen 3.8 Max | max | 0.0% |