List your company

Proactive Assistance on Pro2Bench HowTo

88G-Mean F1Qwen3.6-VL-27B
-3.5220.244467.76Jun 3, 2026

Evaluation results

MethodLinks
Qwen3.6-VL-27BFT=true, PWR=true, Architecture=split duplex+planner2026.068842
Llama 4-MaverickFT=true, PWR=true, Architecture=split duplex+planner2026.067673
Llama 4-MaverickFT=true, Architecture=monolithic2026.066948
Qwen3-VL-235BArchitecture=monolithic2026.064933
Gemini 3.1 ProArchitecture=monolithic, Mode=zero-shot2026.064835
Claude Opus 4.6Architecture=monolithic, Mode=zero-shot2026.064627
Qwen3.6-VL-27BFT=true, Architecture=monolithic2026.064127
Qwen3.6-VL-27BArchitecture=monolithic2026.063341
Llama 4-MaverickArchitecture=monolithic2026.062124
GPT-5.2Architecture=monolithic, Mode=zero-shot2026.060—