Skip to content

Benchmark

QwenReactBench

Scores are Elo ratings. Higher ratings are better.

Models measured
5
Scores recorded
6
Sources
2
IQ weight
0.23

Score distribution

1,669
Mean
1,724
Median
86.7
Std dev
1,564
10th percentile
1,770
90th percentile
1,538
Min
1,5381,770
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 0.23 for chat models, measured over 5 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
1Claude Fable 5Qwen内部双语 React 基准 · Claude Code harness · 自动渲染 + 多模态评判 · BT/Elo· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.1,770
2Qwen3.8 Max· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim1,724
3Qwen3.8 MaxQwen内部双语 React 基准 · Claude Code harness · 自动渲染 + 多模态评判 · BT/Elo· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.1,724
4Claude Opus 4.8Qwen内部双语 React 基准 · Claude Code harness · 自动渲染 + 多模态评判 · BT/Elo· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.1,694
5GPT-5.6 SolQwen内部双语 React 基准 · Claude Code harness · 自动渲染 + 多模态评判 · BT/Elo· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.1,564
6Qwen3.7 MaxQwen内部双语 React 基准 · Claude Code harness · 自动渲染 + 多模态评判 · BT/Elo· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.1,538

Data credits: BenchBoard by SYuan03 (github.com/SYuan03/benchboard). MIT covers its code and data organization; no results licence is inferred. Numeric relays retain the original stating-source links. · Data by LLM Stats