Gemini 2.5 Pro Scores 100 to Take First Place: YZ Index Smoke Quick Test Briefing for July 13, 2026

The July 13, 2026 YZ Index Smoke Quick Test covered 10 models, with Gemini 2.5 Pro scoring a perfect 100 to take first place for the day. Smoke is a daily 10-question quick test suitable for observing short-term signals; it does not equate to the Full weekly ranking conclusion.

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

Daily Rankings

Data Interpretation

In today's YZ Index Smoke Quick Test, Gemini 2.5 Pro achieved a perfect 100 on the main board with Code Execution at 100 and Material Constraints at 100, demonstrating balanced high performance across both dimensions. GPT-o3 and Grok 4 tied for second with Code Execution at 97 and Material Constraints at 97.8, yielding a main board score of 97.36. Doubao Pro and Gemini 3.1 Pro tied at main board 96.7 with Code Execution at 94 and Material Constraints at 100, indicating a complementary strength pattern among top models. Claude Opus 4.7 showed Code Execution at 69 and Material Constraints at 100, main board 82.95; Claude Sonnet 4.6 had Code Execution at 94 and Material Constraints at 62, main board 79.6. Such structural differences highlight models' relative strengths or weaknesses in individual dimensions.

Regarding anomaly signals, Qwen3 Max's main board score of 72 dropped 22.2 points from the previous test, with Code Execution and Material Constraints both falling to 72. Claude Sonnet 4.6's main board dropped 17.4 points to 79.6, with Material Constraints plunging 31.3 points to 62. DeepSeek V4 Pro's main board fell 16.9 points to 80.1, with Code Execution down 28 points to 72. Claude Opus 4.7's main board dropped 14 points to 82.95. These changes could stem from question sampling fluctuations or reflect genuine single-day performance differences, requiring subsequent runs for verification. GLM-4.6 did not participate in the ranking due to missing data in the communication dimension.

Overall, as a small-sample single-day signal, the Smoke Quick Test's score structure for top models and anomalous models provides a daily observation perspective. However, any interpretation should remain cautious to avoid long-term inferences.

Key Changes

• Qwen3 Max: Main board down 22.2 points, Code Execution -23, Material Constraints -21.3
• Claude Sonnet 4.6: Main board down 17.4 points, Code Execution -6, Material Constraints -31.3
• GPT-o3: Main board up 17.1 points, Code Execution +22, Material Constraints +11.1
• DeepSeek V4 Pro: Main board down 16.9 points, Code Execution -28
• Gemini 3.1 Pro: Main board up 16.2 points, Code Execution +24, Material Constraints +6.7

Signals to Watch

• Claude Opus 4.7: Main board plummeted -14 points
• DeepSeek V4 Pro: Main board plummeted -16.9 points
• Claude Sonnet 4.6: Main board plummeted -17.4 points
• Qwen3 Max: Main board plummeted -22.2 points
• GLM-4.6: Incomplete data (missing communication dimension, API failure/timeout); automatic re-run initiated; not ranked in this round

When reading such Smoke briefings, the focus should be on two questions: first, whether a model has consistently exposed the same weakness over consecutive days; second, whether the integrity rating has shifted from pass to warn or fail. Large daily changes in execution or constraint scores may stem from question sampling or be early signals of genuine degradation, requiring subsequent runs for verification.

Data Source: YZ Index | Run #229 | View Raw Data