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

Progress Reasoning on Progress-Bench Same-View 1.0 (test)

10.3NSE

ProgressLM-3B-RL

Updated 3mo ago

Evaluation Results

Method	Links
ProgressLM-3B-RL 2026.01		10.3	93.5	0.1
GPT-5 2026.01		14.6	89.4	0
ProgressLM-3B-SFT 2026.01		15.5	84	0.6
Qwen3-VL-32B 2026.01		15.9	88.3	0
Qwen2.5-VL-72B 2026.01		16.8	83.9	0.4
Qwen3-VL-8B 2026.01		19.1	81.7	0
GPT-5-mini 2026.01		19.8	84.1	0.4
Qwen2.5-VL-32B 2026.01		21.4	72.9	0
Qwen3-VL-4B 2026.01		21.6	72.3	0
Qwen2.5-VL-7B 2026.01		27.3	51.9	30.6
Intern3.5-VL-38B 2026.01		27.6	74.8	0.5
Qwen2.5-VL-3B 2026.01		29.2	43	9.9
Intern3.5-VL-4B 2026.01		44.4	33.1	0.7
Qwen3-VL-2B 2026.01		60	35	0.1
Intern3.5-VL-14B 2026.01		62.1	24.2	0
Intern3.5-VL-8B 2026.01		65.2	5.2	0.5