On 2026-10-16, the YZ Index Smoke Quick Test covered 13 models, with Doubao Pro ranking first for the day at 85.09. Smoke is a daily 10-question quick test suited to observing short-term signals and is not equivalent to Full weekly ranking conclusions.
This Smoke evaluation covers only two Main Ranking dimensions: Code Execution and Material Constraints. The Main Ranking formula is 0.55 × Code Execution + 0.45 × Material Constraints. Because the daily sample size is small, a single-day score is better treated as a monitoring signal rather than a long-term verdict on model capability.
Daily Ranking
| Rank | Model | Main Ranking | Code Execution | Material Constraints | Integrity |
|---|---|---|---|---|---|
| #1 | Doubao Pro | 85.09 | 95.3 | 72.6 | pass |
| #2 | Grok 4 | 80.81 | 95.3 | 63.1 | pass |
| #3 | DeepSeek V4 Pro | 79.25 | 72 | 88.1 | pass |
| #4 | GPT-6 Astra | 79.25 | 72 | 88.1 | pass |
| #5 | GPT-6.1 Sol | 79.25 | 72 | 88.1 | pass |
| #6 | Claude Opus 4.7 | 73.45 | 68.9 | 79 | pass |
| #7 | GPT-6 Luna | 71.51 | 69.8 | 73.6 | pass |
| #8 | Gemini 3.1 Pro | 69.62 | 54.5 | 88.1 | pass |
| #9 | GPT-5.5 | 62.48 | 47 | 81.4 | pass |
| #10 | Gemini 2.5 Pro | 62.42 | 50 | 77.6 | pass |
| #11 | GPT-o3 | 60.71 | 38.3 | 88.1 | pass |
| #12 | Claude Sonnet 4.6 | 59.56 | 44.8 | 77.6 | pass |
| #13 | GPT-6 Sol | 53.04 | 44.8 | 63.1 | pass |
Data Interpretation
In today's YZ Index Smoke Quick Test, Doubao Pro ranked first with a Main Ranking score of 85.09. Its Code Execution score of 95.3 and Material Constraints score of 72.6 form a structure of high Code Execution paired with moderate Material Constraints; Grok 4 also scored 95.3 in Code Execution but only 63.1 in Material Constraints, following closely with 80.81 on the Main Ranking. DeepSeek V4 Pro, GPT-6 Astra, and GPT-6.1 Sol all scored 72 in Code Execution and 88.1 in Material Constraints, tying at 79.25 on the Main Ranking, showing a stronger Material Constraints profile. Claude Opus 4.7 scored 68.9 in Code Execution and 79 in Material Constraints, with 73.45 on the Main Ranking; Gemini 3.1 Pro scored 54.5 in Code Execution and 88.1 in Material Constraints, with 69.62 on the Main Ranking. Such models complement the leaders in the Material Constraints dimension.
Several models showed significant swings: GPT-6 Sol fell 34.8 points on the Main Ranking, 47.2 points in Code Execution, and 19.7 points in Material Constraints; Gemini 2.5 Pro fell 33.5 points on the Main Ranking and 47.8 points in Code Execution; Claude Sonnet 4.6 fell 31.2 points on the Main Ranking and 48.2 points in Code Execution. GPT-5.5 fell 25.4 points on the Main Ranking and 45 points in Code Execution; GPT-o3 fell 23.9 points on the Main Ranking, 33.7 points in Code Execution, and 11.9 points in Material Constraints. Doubao Pro's Material Constraints plunged 15.3 points, Claude Opus 4.7's Main Ranking plunged 20.1 points, and GPT-6 Luna's Main Ranking plunged 14.5 points. These changes may stem from question sampling fluctuations or may reflect real degradation, and require follow-up runs for confirmation. GLM-4.6 and Qwen3 Max did not participate in the ranking due to incomplete data.
The Smoke quick test is a small-sample, single-day signal. The observations above reflect only the day's score structure and swings and should not be used for long-term inference.
Key Changes
- GPT-6 Sol: Main Ranking down 34.8 points, Code Execution -47.2 points, Material Constraints -19.7 points
- Gemini 2.5 Pro: Main Ranking down 33.5 points, Code Execution -47.8 points, Material Constraints -16 points
- Claude Sonnet 4.6: Main Ranking down 31.2 points, Code Execution -48.2 points, Material Constraints -10.3 points
- GPT-5.5: Main Ranking down 25.4 points, Code Execution -45 points
- GPT-o3: Main Ranking down 23.9 points, Code Execution -33.7 points, Material Constraints -11.9 points
Signals to Watch
- Doubao Pro: Material Constraints plunged -15.3 points
- Claude Opus 4.7: Main Ranking plunged -20.1 points
- GPT-6 Luna: Main Ranking plunged -14.5 points
- GPT-5.5: Main Ranking plunged -25.4 points
- Gemini 2.5 Pro: Main Ranking plunged -33.5 points
- GPT-o3: Main Ranking plunged -23.9 points
- Claude Sonnet 4.6: Main Ranking plunged -31.2 points
- GPT-6 Sol: Main Ranking plunged -34.8 points
- GLM-4.6: Incomplete data (missing Integrity and Communication dimensions; API failure/timeout); has entered automatic rerun and does not participate in this period's ranking
- Qwen3 Max: Incomplete data (missing multiple evaluation dimensions, including Execution, Judgment, Integrity, and Communication; API failure/timeout); has entered automatic rerun and does not participate in this period's ranking
When reading this type of Smoke brief, focus on two questions: first, whether a model has exposed the same type of weakness for multiple consecutive days; second, whether its Integrity rating has moved from pass to warn or fail. Large single-day changes in Execution or Constraints scores may come from question sampling or may be an early signal of real degradation, requiring follow-up runs for confirmation.
Data source: YZ Index | Run #381 | View raw data
© 2026 Winzheng.com 赢政天下 | 转载请注明来源并附原文链接