Qwen-72B-Chat: benchmark results
Alibaba Cloud's Qwen-72B-Chat, the aligned chat version of Qwen-72B, tuned to follow system instructions for role-play, style transfer and task-specific behavior, with 32K context. Provider: Alibaba. Released 2023-11-30. Access: Open.
Aggregate Rating 1482 ± 18, rank #835 of 1635 rated models, from 103 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| OpenCompass Knowledge - Humanities - Chinese (CompassBench 2404) | 59.4 | Score (%) | 100 |
| OpenCompass Knowledge - Natural Science - Chinese (CompassBench 2404) | 58.4 | Score (%) | 100 |
| OpenCompass Knowledge - Social Science - Chinese (CompassBench 2404) | 73.1 | Score (%) | 100 |
| OpenCompass LLM - Knowledge - Chinese (CompassBench 2404) | 66.8 | Score (%) | 100 |
| OpenCompass Language - Content Summarization - English (CompassBench 2404) | 36.8 | Score (%) | 100 |
| OpenCompass Language - Translation (CompassBench 2404) | 22.4 | Score (%) | 100 |
| OpenCompass Knowledge - Common Sense - Chinese (CompassBench 2404) | 76.5 | Score (%) | 96.7 |
| OpenCompass LLM - Knowledge (CompassBench 2404) | 71.7 | Score (%) | 96.7 |
| OpenCompass Math - Middle - Chinese (CompassBench 2404) | 53.4 | Score (%) | 93 |
| OpenCompass LLM - Knowledge - English (CompassBench 2404) | 76.7 | Score (%) | 90 |
| OpenCompass Language - Chinese Semantic Understanding - Chinese (CompassBench 2404) | 63.7 | Score (%) | 90 |
| OpenCompass Reasoning - Deductive - English (CompassBench 2404) | 20.9 | Score (%) | 88.3 |
Interactive version: theaggregate.ai/model?slug=qwen-72b-chat · How It Works · Data refreshed daily, snapshot 2026-11-05.