Qwen-7B Chat: benchmark results

Alibaba Cloud's Qwen-7B-Chat, the aligned chat version of Qwen-7B, an updated release building on Alibaba's earlier Qwen-7B checkpoint. Provider: Alibaba. Released 2023-08-03. Access: Open.

Aggregate Rating 1353 ± 14, rank #1380 of 1656 rated models, from 113 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
OpenCompass Language - Content Summarization - Chinese (CompassBench 2404)28.6Score (%)83.3
OpenCompass Language - Translation (CompassBench 2404)16.3Score (%)75
OpenCompass Language - Sentiment Analysis - Chinese (CompassBench 2404)88Score (%)70
OpenCompass Language - Sentiment Analysis - English (CompassBench 2404)68.7Score (%)68.3
OpenCompass Reasoning - Inductive - Chinese (CompassBench 2404)30.2Score (%)68.3
OpenCompass Agent - CIBench - English (CompassBench 2404)42Score (%)66.7
OpenCompass Reasoning - Common Sense - English (CompassBench 2404)20.2Score (%)66.7
OpenCompass Agent - CIBench - English (CompassBench 2402)49.4Score (%)64.7
OpenCompass Reasoning - Common Sense - Chinese (CompassBench 2404)21.1Score (%)63.3
OpenCompass Language - Intention Recognition - Chinese (CompassBench 2404)41.3Score (%)61.7
OpenCompass Agent - T-Eval - Chinese (CompassBench 2404)61.1Score (%)60
OpenCompass Knowledge - Humanities - Chinese (CompassBench 2404)39Score (%)60

Interactive version: theaggregate.ai/model?slug=qwen-7b-chat · How It Works · Data refreshed daily, snapshot 2026-10-20.