The 2026-07-15 YZ Index Smoke quick test covered 10 models. Claude Opus 4.7, Gemini 2.5 Pro, and GPT-5.5 tied at 100 points for the top spot of the day. Smoke is a daily 10-question quick test, suitable for observing short-term signals, and is not equivalent to the Full weekly ranking conclusions.
This Smoke evaluation only covers two main dimensions: Code Execution and Material Constraint. The overall score formula is 0.55 × Code Execution + 0.45 × Material Constraint. Due to the small daily sample size, a single day's score is more suitable as a monitoring signal rather than a long-term conclusion on model capabilities.
Daily Ranking
| Rank | Model | Overall Score | Code Execution | Material Constraint | Integrity |
|---|---|---|---|---|---|
| #1 | Claude Opus 4.7 | 100 | 100 | 100 | pass |
| #2 | Gemini 2.5 Pro | 100 | 100 | 100 | pass |
| #3 | GPT-5.5 | 100 | 100 | 100 | pass |
| #4 | Claude Sonnet 4.6 | 88.75 | 100 | 75 | pass |
| #5 | GPT-o3 | 88.75 | 100 | 75 | pass |
| #6 | DeepSeek V4 Pro | 86.25 | 75 | 100 | pass |
| #7 | Doubao Pro | 86.25 | 75 | 100 | pass |
| #8 | Grok 4 | 83.67 | 70.3 | 100 | pass |
| #9 | Qwen3 Max | 80.17 | 84.4 | 75 | pass |
| #10 | Gemini 3.1 Pro | 72.5 | 50 | 100 | pass |
Key Changes
- Claude Sonnet 4.6: Overall Score up 27.7 points, Code Execution +25, Material Constraint +30.9
- Qwen3 Max: Overall Score up 17.8 points, Code Execution +25.9, Material Constraint +7.8
- Gemini 2.5 Pro: Overall Score up 13.9 points, Material Constraint +30.9
- Claude Opus 4.7: Overall Score up 13.8 points, Code Execution +25
- Grok 4: Overall Score up 8.8 points, Material Constraint +25.2, Integrity warn→pass
Signals to Watch
- DeepSeek V4 Pro: Code Execution plummeted by 23.7 points
- Gemini 3.1 Pro: Overall Score plummeted by 8.5 points
- GLM-4.6: Incomplete data (missing judgment, communication dimensions, API failure/timeout). Has entered automatic re-run, not participating in this period's ranking.
When reading these Smoke briefings, the focus should be on two questions: First, whether a model has continuously exposed the same type of weakness over multiple days; Second, whether the integrity rating drops from pass to warn or fail. Large daily changes in execution or constraint scores may come from question sampling, or may be early signals of genuine degradation, requiring subsequent run verification.
Data source: YZ Index | Run #232 | View raw data
© 2026 Winzheng.com 赢政天下 | 转载请注明来源并附原文链接