Doubao Pro Leads with 91.11 Points: 2026-11-21 Smoke Quick-Test Data Brief

The 2026-11-21 YZ Index Smoke quick test covered 13 models, with Doubao Pro ranking first for the day at 91.11 points. 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 covered only the 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 capabilities.

Daily Rankings

RankModelMain LeaderboardCode ExecutionMaterial ConstraintsIntegrity
#1Doubao Pro91.1198.881.7pass
#2Grok 480.5898.858.3pass
#3GPT-6 Luna80.1399.356.7pass
#4GPT-o377.6198.851.7pass
#5Gemini 3.1 Pro72.757570pass
#6Claude Opus 4.772.3774.370pass
#7GPT-6.1 Sol72.3774.370pass
#8Gemini 2.5 Pro69.747563.3pass
#9GPT-6 Astra69.3574.363.3pass
#10GPT-6 Sol69.3574.363.3pass
#11Claude Sonnet 4.669.0873.863.3pass
#12DeepSeek V4 Pro67.497558.3pass
#13GPT-5.555.6549.363.4pass

Key Changes

  • GPT-6 Luna: Main leaderboard up 24.2 points; code execution +54.7 points; material constraints -13.1 points
  • Claude Opus 4.7: Main leaderboard down 19.8 points; code execution -20.2 points; material constraints -19.3 points
  • GPT-o3: Main leaderboard up 19.7 points; code execution +53.3 points; material constraints -21.4 points
  • GPT-6 Astra: Main leaderboard down 19.6 points; code execution -22.5 points; material constraints -16 points
  • Gemini 3.1 Pro: Main leaderboard up 17.7 points; code execution +45.5 points; material constraints -16.3 points

Signals to Watch

  • No anomaly signals suitable for publication were retained in this run.

When reading this kind of Smoke brief, focus on two questions: first, whether a model has shown the same type of weakness for several consecutive 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 early signals of real degradation, and should be tracked in subsequent runs.


Data source: YZ Index | Run #437 | View raw data

This article is from Winzheng Index blog, translated in full by Winzheng (winzheng.com). Click here to view the original When republishing the translation, please credit the source. Thank you!