Share your thoughts, 1 month free Claude Pro on usSee more

Proactive Assistance on Pro2Bench EK

0.9G-Mean F1

Qwen3.6-VL-27B

Updated 1mo ago

Evaluation Results

Method	Links
Qwen3.6-VL-27B 2026.06		0.9	0.36
Llama 4-Maverick 2026.06		0.8	0.67
Llama 4-Maverick 2026.06		0.71	0.51
Claude Opus 4.6 2026.06		0.52	0.29
Gemini 3.1 Pro 2026.06		0.5	0.34
Llama 4-Maverick 2026.06		0.5	0.36
Qwen3.6-VL-27B 2026.06		0.47	0.3
Qwen3-VL-235B 2026.06		0.46	0.32
Qwen3.6-VL-27B 2026.06		0.17	0.45
GPT-5.2 2026.06		0	-