SEQUOR Everything: leaderboard
Metric: Mixed constraint operations: last-minus-first-turn accuracy change (percentage points) (self-reported). Source: benchmarklist.com. Saturation forecast: Around 2029. 11 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3.1 Flash Lite | -9 |
| 2 | GLM-4.7 Flash | -16 |
| 3 | Llama 3.3 70B Instruct | -22 |
| 4 | Gemma 3 12B | -27 |
| 5 | Gemma 3 27B | -30.5 |
| 6 | GPT-OSS-20B | -32 |
| 7 | GPT-OSS-120B | -41 |
| 8 | Gemma 3 4B | -48 |
Interactive version: theaggregate.ai/benchmark?slug=sequor-everything · How It Works · Data refreshed daily, snapshot 2026-10-09.