SEQUOR Tuples: leaderboard
Metric: Initial constraint tuple: last-minus-first-turn accuracy change (percentage points) (self-reported). Source: benchmarklist.com. Saturation forecast: Around January 2028. 11 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GLM-4.7 Flash | -16.5 |
| 2 | GPT-OSS-20B | -23.5 |
| 3 | Gemini 3.1 Flash Lite | -24.42 |
| 4 | GPT-OSS-120B | -41.5 |
| 5 | Llama 3.3 70B Instruct | -44.5 |
| 6 | Gemma 3 12B | -54 |
| 7 | Gemma 3 27B | -54.5 |
| 8 | Gemma 3 4B | -62 |
Interactive version: theaggregate.ai/benchmark?slug=sequor-tuples · How It Works · Data refreshed daily, snapshot 2026-10-09.