Qwen 1.5 7B Chat: benchmark results
Alibaba's Qwen1.5-7B-Chat, the aligned chat version in the Qwen1.5 (Qwen2 beta) series, post-trained with SFT and DPO, supporting a stable 32K context. Provider: Alibaba. Released 2024-02-04. Access: Open.
Aggregate Rating 1374 ± 10, rank #1324 of 1656 rated models, from 290 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| RewardBench Reasoning | 90.41 | Accuracy (%) | 74 |
| OpenCompass Agent - CIBench - English (CompassBench 2402) | 51.1 | Score (%) | 70.6 |
| ChineseSafe Benchmark | 70.36 | Accuracy (%) | 70.4 |
| Open Portuguese LLM - FaQuAD NLI | 72.46 | Macro F1 (%) | 65.7 |
| OpenCompass Math - Middle School - Chinese (CompassBench 2405) | 47.4 | Score (%) | 64.5 |
| BiGGen-Bench | 3.56 | Average Score (1-5) | 63.7 |
| RewardBench Chat Hard | 69.08 | Accuracy (%) | 63.6 |
| OpenCompass Language - Content Summarization - English (CompassBench 2405) | 28.8 | Score (%) | 63.2 |
| MERA - RWSD | 58.46 | Accuracy (%) | 62.6 |
| Open PL LLM - PolEmo2-OUT (multiple choice, 5-shot) | 71.26 | Accuracy (%) | 62.5 |
| OpenCompass Subjective - Creation (CompassBench 2404) | 38.3 | Score (%) | 62 |
| Open LLM Leaderboard - GPQA | 30.29 | Score | 59.3 |
Interactive version: theaggregate.ai/model?slug=qwen-1-5-7b-chat · How It Works · Data refreshed daily, snapshot 2026-10-20.