Claude Opus 4.7 Tops with 90.51 Points: 2026-07-09 Smoke Quick Test Data Brief

On July 9, 2026, the YZ Index Smoke Quick Test covered 10 models, with Claude Opus 4.7 ranking first that day with a score of 90.51. Smoke is a daily 10-question quick test, suitable for observing short-term signals, and is not equivalent to the conclusions of the Full weekly ranking.

This Smoke evaluation only covers two main ranking dimensions: code execution and material constraint. The main ranking formula is 0.55 × Code Execution + 0.45 × Material Constraint. Due to the small daily sample size, single-day scores are better used as monitoring signals rather than as long-term conclusions about model capabilities.

Daily Ranking

Data Interpretation

In today's Smoke Quick Test, Claude Opus 4.7 topped the main ranking at 90.51, with a balanced combination of Code Execution at 91.5 and Material Constraint at 89.3. Gemini 3.1 Pro scored 89.16 on the main ranking, with Code Execution also at 91.5 and Material Constraint at 86.3, slightly weaker than the leader on the material side. DeepSeek V4 Pro scored 88.13 on the main ranking, with Code Execution at 91.5 and Material Constraint at 84, structurally similar to the top two. Grok 4 scored 87.66 on the main ranking, with Code Execution relatively high at 94.5 and Material Constraint at 79.3, showing a clear contrast of strong code and weak materials.

Gemini 3.1 Pro saw a +28.2 point increase in the main ranking, +33.2 points in Code Execution, and +22 points in Material Constraint. Gemini 2.5 Pro saw a +24.3 point increase in the main ranking, +34.3 points in Code Execution, while 豆包 Pro saw a +12.8 point increase in the main ranking and +18.5 points in Material Constraint. The increases are concentrated in synchronous or biased improvements in Code Execution and Material Constraint. Qwen3 Max suffered a sharp drop of -15.1 points in Material Constraint, and GPT-o3 dropped sharply by -16.8 points in Material Constraint. These anomalies may come from question sampling fluctuations or could indicate real degradation, requiring follow-up runs for verification. GLM-4.6 was not included in the ranking due to incomplete data.

Key Changes

• Gemini 3.1 Pro: Main ranking up 28.2 points, Code Execution +33.2 points, Material Constraint +22 points
• Gemini 2.5 Pro: Main ranking up 24.3 points, Code Execution +34.3 points, Material Constraint +12 points
• 豆包 Pro: Main ranking up 12.8 points, Code Execution +8.2 points, Material Constraint +18.5 points
• GPT-5.5: Main ranking up 10.8 points, Code Execution +16.5 points
• Claude Opus 4.7: Main ranking up 9.1 points, Code Execution +16.5 points

Signals to Watch

• Qwen3 Max: Material Constraint dropped sharply by -15.1 points
• GPT-o3: Material Constraint dropped sharply by -16.8 points
• GLM-4.6: Data incomplete (missing grounding, judgment, integrity dimensions; API failure/timeout), has entered automatic retry; not included in this period's ranking

When reading such Smoke briefings, the focus should be on two questions: first, whether a model has exposed the same type of weakness for multiple consecutive days; second, whether the integrity rating has moved from pass to warn or fail. Large daily fluctuations in execution or constraint scores may be due to question sampling or could be early signals of real degradation, requiring follow-up runs for verification.

Data source: YZ Index | Run #222 | View raw data