CMDB-1500 (Text): leaderboard

Surd AI CMDB-1500 fixed-input decision benchmark: accuracy on 1200 text questions spanning candidate choice, yes/no, ordinal and multi-select decisions. The vendor-run leaderboard includes general-purpose LLMs and Surd AI decision models; text and image panels use separate fixed denominators.

Metric: Accuracy (%), fixed denominator of 1,200 questions (invalid answers count as wrong), effort as labelled. Source: surdai.com. Saturation forecast: Around March 2027. 10 models tracked. Updated 7 Oct 2026.

Top models

#ModelSettingScoreOverall rank
1GPT-6.1 SolLow80#15
2SPX-CD-Pro79.33
3Claude Sonnet 5.5Low78#24
4SPX-CD-Flash77.58
5GPT-6 LunaLow76.58#105
6GLM-5.3 FlashLow75.75#133
7SPX-CD-Omni74.67
8Qwen 3.8 Flash71.08#125
9Kimi K2.671#192
10DeepSeek V4.1 Flash68.58#67

No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=cmdb-1500-text · How It Works · Data refreshed daily, snapshot 2026-11-21.