abab6.5s-chat: benchmark results
Provider: MiniMax. Access: API.
Aggregate Rating 1503 ± 30, rank #2289 of 10916 rated models, from 47 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| OpenCompass Reasoning - Common Sense - English (CompassBench 2409) | 47.2 | Score (%) | 69 |
| OpenCompass Reasoning - Common Sense - Chinese (CompassBench 2409) | 48.7 | Score (%) | 65.5 |
| OpenCompass Knowledge - Engineering - English (CompassBench 2409) | 82 | Score (%) | 51.7 |
| OpenCompass Language - Dialogue - Chinese (CompassBench 2409) | 46.2 | Score (%) | 51.7 |
| OpenCompass Knowledge - Engineering - Chinese (CompassBench 2409) | 72 | Score (%) | 48.3 |
| OpenCompass Language - Creation - Chinese (CompassBench 2409) | 41.8 | Score (%) | 48.3 |
| OpenCompass Reasoning - Science and Engineering - English (CompassBench 2409) | 43.6 | Score (%) | 48.3 |
| OpenCompass Code - HumanEval+ - English (CompassBench 2409) | 72 | Score (%) | 44.8 |
| OpenCompass Code - LCBench - English (CompassBench 2409) | 37.8 | Score (%) | 44.8 |
| OpenCompass LLM - Knowledge (CompassBench 2409) | 69.9 | Score (%) | 44.8 |
| OpenCompass LLM - Knowledge - English (CompassBench 2409) | 76.8 | Score (%) | 44.8 |
| OpenCompass Math - College - English (CompassBench 2409) | 37.3 | Score (%) | 44.8 |
Interactive version: theaggregate.ai/model?slug=abab6-5s-chat · How It Works · Data refreshed daily, snapshot 2026-11-08.