SEQUOR: leaderboard
Metric: Single constraint: last-minus-first-turn accuracy change (percentage points) (self-reported). Source: benchmarklist.com. Saturation forecast: Around February 2028. 11 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GLM-4.7 Flash | -4.5 |
| 2 | Gemini 3.1 Flash Lite | -10.5 |
| 3 | Llama 3.3 70B Instruct | -30 |
| 4 | Gemma 3 12B | -34 |
| 5 | GPT-OSS-20B | -34.5 |
| 6 | GPT-OSS-120B | -35.5 |
| 7 | Gemma 3 27B | -40 |
| 8 | Gemma 3 4B | -54.5 |
Interactive version: theaggregate.ai/benchmark?slug=sequor · How It Works · Data refreshed daily, snapshot 2026-10-09.