The 2026-11-06 YZ Index Smoke quick test covered 13 models, with Doubao Pro ranking first for the day at 87.6 points. Smoke is a daily 10-question quick test; it is suitable for observing short-term signals and is not equivalent to the conclusions of the Full weekly ranking.
This Smoke evaluation covers only the two main leaderboard dimensions: code execution and material constraints. The main leaderboard formula is 0.55 × code execution + 0.45 × material constraints. Because the daily sample size is small, single-day scores are better used as monitoring signals rather than long-term conclusions about model capability.
Daily ranking
| Rank | Model | Main leaderboard | Code execution | Material constraints | Integrity |
|---|---|---|---|---|---|
| #1 | Doubao Pro | 87.6 | 90.3 | 84.3 | pass |
| #2 | Claude Opus 4.7 | 80.66 | 69.5 | 94.3 | pass |
| #3 | Gemini 3.1 Pro | 79.69 | 74.2 | 86.4 | pass |
| #4 | GPT-6 Astra | 79.29 | 94.5 | 60.7 | pass |
| #5 | GPT-6.1 Sol | 79.29 | 94.5 | 60.7 | pass |
| #6 | Grok 4 | 74.26 | 82 | 64.8 | pass |
| #7 | GPT-5.5 | 73.91 | 69.5 | 79.3 | pass |
| #8 | Gemini 2.5 Pro | 73.85 | 65.3 | 84.3 | pass |
| #9 | DeepSeek V4 Pro | 73.46 | 69.5 | 78.3 | pass |
| #10 | Claude Sonnet 4.6 | 69.76 | 82 | 54.8 | pass |
| #11 | GPT-6 Sol | 68.11 | 69.5 | 66.4 | pass |
| #12 | GPT-o3 | 67.79 | 69.5 | 65.7 | warn |
| #13 | GPT-6 Luna | 64.13 | 50 | 81.4 | pass |
Key changes
- Claude Opus 4.7: main leaderboard rose 29.6 points, code execution +50.1 points
- GPT-5.5: main leaderboard rose 19.8 points, code execution +44.5 points, material constraints -10.5 points
- Doubao Pro: main leaderboard rose 17.4 points, code execution +40.3 points, material constraints -10.5 points
- Gemini 2.5 Pro: main leaderboard rose 16.2 points, code execution +34 points, material constraints -5.5 points
- GPT-o3: main leaderboard rose 13.6 points, code execution +44.5 points, material constraints -24.1 points, integrity pass→warn
Signals to watch
- GPT-6 Astra: material constraints plunged -29.1 points
- GPT-6.1 Sol: material constraints plunged -29.1 points
- Grok 4: material constraints plunged -25 points
- Claude Sonnet 4.6: material constraints plunged -38 points
- GPT-o3: material constraints plunged -24.1 points
- GLM-4.6: data incomplete (several dimensions unavailable due to API failure/timeout); automatic rerun has been triggered, and it is not included in this period's ranking
- Qwen3 Max: data incomplete (multiple dimensions unavailable due to API failure/timeout); automatic rerun has been triggered, and it is not included in this period's ranking
When reading this kind of Smoke brief, focus on two questions: first, whether a model has shown the same type of weakness for several consecutive days; second, whether its integrity rating has moved from pass to warn or fail. Large single-day changes in execution or constraint scores may come from question sampling, or they may be early signals of real degradation and require follow-up runs for review.
Data: YZ Index | Run #414 | View raw data
© 2026 Winzheng.com 赢政天下 | 转载请注明来源并附原文链接