abab5.5: benchmark results
Provider: MiniMax. Access: API.
Aggregate Rating 1454 ± 25, rank #3575 of 10916 rated models, from 65 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| OpenCompass Reasoning - Common Sense - English (CompassBench 2404) | 37.2 | Score (%) | 100 |
| OpenCompass Reasoning - Deductive - English (CompassBench 2404) | 27.9 | Score (%) | 100 |
| OpenCompass Language - Sentiment Analysis - English (CompassBench 2404) | 78 | Score (%) | 98.3 |
| OpenCompass LLM - Reasoning - English (CompassBench 2404) | 45 | Score (%) | 96.7 |
| OpenCompass Language - Sentiment Analysis - Chinese (CompassBench 2404) | 90 | Score (%) | 96.7 |
| OpenCompass Agent - CIBench - Chinese (CompassBench 2404) | 62.2 | Score (%) | 93.3 |
| OpenCompass Agent - CIBench - English (CompassBench 2404) | 64.5 | Score (%) | 93.3 |
| OpenCompass LLM - Language - English (CompassBench 2404) | 56.8 | Score (%) | 90 |
| OpenCompass Language - Content Summarization - English (CompassBench 2404) | 33.9 | Score (%) | 90 |
| OpenCompass Language - Intention Recognition - English (CompassBench 2404) | 79.3 | Score (%) | 90 |
| OpenCompass LLM - Agent (CompassBench 2404) | 62.2 | Score (%) | 86.7 |
| OpenCompass LLM - Agent - Chinese (CompassBench 2404) | 60.3 | Score (%) | 86.7 |
Interactive version: theaggregate.ai/model?slug=abab5-5 · How It Works · Data refreshed daily, snapshot 2026-11-08.