Nanbeige2-8B-Chat: benchmark results
Provider: Nanbeige.
Aggregate Rating 1412 ± 27, rank #5539 of 10916 rated models, from 152 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| OpenCompass Agent - T-Eval - Chinese (CompassBench 2405) | 89.6 | Score (%) | 97.4 |
| OpenCompass Agent - T-Eval - Chinese (CompassBench 2404) | 77.5 | Score (%) | 90 |
| OpenCompass Reasoning - Inductive - English (CompassBench 2404) | 34.3 | Score (%) | 88.3 |
| OpenCompass Agent - T-Eval - English (CompassBench 2405) | 86 | Score (%) | 84.2 |
| OpenCompass LLM - Agent - Chinese (CompassBench 2404) | 59.9 | Score (%) | 83.3 |
| OpenCompass LLM - Agent (CompassBench 2404) | 58.7 | Score (%) | 80 |
| OpenCompass Language - Content Critique - Chinese (CompassBench 2404) | 48 | Score (%) | 80 |
| OpenCompass Agent - T-Eval - English (CompassBench 2404) | 74.3 | Score (%) | 76.7 |
| OpenCompass Agent - CIBench - Chinese (CompassBench 2404) | 42.4 | Score (%) | 73.3 |
| OpenCompass LLM - Agent - English (CompassBench 2404) | 57.4 | Score (%) | 70 |
| OpenCompass Subjective - Creation (CompassBench 2404) | 38.5 | Score (%) | 64 |
| OpenCompass Code - MBPP - Chinese (CompassBench 2404) | 38 | Score (%) | 63.3 |
Interactive version: theaggregate.ai/model?slug=nanbeige2-8b-chat · How It Works · Data refreshed daily, snapshot 2026-11-08.