The 2026-11-15 YZ Index Smoke quick test covered 13 models, with Claude Sonnet 4.6 and GPT-6 Luna tied at 95.19 points for first place that day. Smoke is a daily 10-question quick test; it is suitable for observing short-term signals and is not equivalent to Full weekly leaderboard conclusions.
This Smoke evaluation covered only 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 Rankings
| Rank | Model | Main | Code Execution | Material Constraints | Integrity |
|---|---|---|---|---|---|
| #1 | Claude Sonnet 4.6 | 95.19 | 100 | 89.3 | pass |
| #2 | GPT-6 Luna | 95.19 | 100 | 89.3 | pass |
| #3 | GPT-o3 | 94.31 | 98.4 | 89.3 | pass |
| #4 | GPT-6 Sol | 90.69 | 100 | 79.3 | pass |
| #5 | GPT-5.5 | 88.75 | 100 | 75 | pass |
| #6 | GPT-6 Astra | 84.7 | 97.3 | 69.3 | pass |
| #7 | GPT-6.1 Sol | 84.7 | 97.3 | 69.3 | pass |
| #8 | Grok 4 | 82.7 | 77.3 | 89.3 | pass |
| #9 | Claude Opus 4.7 | 81.44 | 75 | 89.3 | pass |
| #10 | Doubao Pro | 65.38 | 45.8 | 89.3 | pass |
| #11 | Gemini 3.1 Pro | 53.94 | 25 | 89.3 | pass |
| #12 | Gemini 2.5 Pro | 51.63 | 20.8 | 89.3 | pass |
| #13 | DeepSeek V4 Pro | 49.44 | 25 | 79.3 | pass |
Key Changes
- GPT-o3: Main score up 28.8 points, Code Execution +7.1 points, Material Constraints +55.2 points
- GPT-6 Sol: Main score up 24.8 points, Code Execution +8.7 points, Material Constraints +44.5 points
- Claude Sonnet 4.6: Main score up 24.6 points, Code Execution +5.5 points, Material Constraints +47.9 points
- GPT-6 Astra: Main score up 19.6 points, Material Constraints +40.2 points
- GPT-6 Luna: Main score up 18.6 points, Code Execution +5.5 points, Material Constraints +34.5 points
Signals to Watch
- Claude Opus 4.7: Code Execution plunged -19.5 points
- Doubao Pro: Main score plunged -8 points
- Gemini 3.1 Pro: Code Execution plunged -19.5 points
- Gemini 2.5 Pro: Code Execution plunged -46.2 points
- DeepSeek V4 Pro: Code Execution plunged -19.5 points
- GLM-4.6: Data incomplete (missing multiple dimensions, API failure/timeout); automatically queued for rerun and not ranked in this issue
- Qwen3 Max: Data incomplete (missing multiple dimensions, API failure/timeout); automatically queued for rerun and not ranked in this issue
When reading this type of Smoke brief, focus on two questions: first, whether a model has repeatedly exposed the same type of weakness over multiple 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 may be an early signal of real degradation, which requires follow-up runs for confirmation.
Data source: YZ Index | Run #427 | View raw data
© 2026 Winzheng.com 赢政天下 | 转载请注明来源并附原文链接