Multimodal Reasoning on O3-BENCH (test) benchmark leaderboard · SOTA2 Research

Multimodal Reasoning on O3-BENCH (test)

0.756Chart ScoreINSIGHT-O3

Evaluation Results

Method	Links
INSIGHT-O32025.12		0.756	0.644	0.697
Qwen3-VL-32B-Instruct2025.12		0.737	0.485	0.604
Qwen3-VL-235B-A22B-Instruct2025.12		0.734	0.538	0.631
Gemini-3-Flash2025.12		0.681	0.69	0.686
Gemini-3-Pro-Preview2025.12		0.677	0.696	0.687
Gemini-2.5-Pro2025.12		0.673	0.637	0.654
INSIGHT-O32025.12		0.673	0.564	0.615
Gemini-2.5-Flash2025.12		0.618	0.592	0.604
Qwen3-VL-235B-A22B-Thinking2025.12		0.573	0.478	0.523
doubao-seed-1-6-2506152025.12		0.554	0.485	0.518
Qwen3-VL-8B-Instruct2025.12		0.544	0.339	0.436
Qwen3-VL-32B-Thinking2025.12		0.524	0.405	0.461
GLM-4.6V2025.12		0.515	0.385	0.446
Qwen3-VL-30B-A3B-Thinking2025.12		0.511	0.368	0.436
Qwen3-VL-30B-A3B-Instruct2025.12		0.493	0.321	0.402
Qwen3-VL-8B-Thinking2025.12		0.491	0.33	0.406
Gemini-2.5-Flash2025.12		0.466	0.526	0.498
Qwen2.5-VL-32B-Instruct2025.12		0.354	0.335	0.344
INSIGHT-O32025.12		0.353	0.341	0.346
GPT-5-mini2025.12		0.344	0.432	0.39
INSIGHT-O32025.12		0.344	0.383	0.364
GPT-5.22025.12		0.319	0.39	0.357
Qwen2.5-VL-7B-Instruct2025.12		0.309	0.244	0.274
GPT-52025.12		0.309	0.526	0.423
OpenAI o32025.12		0.278	0.524	0.408
InternVL3.5-8B2025.12		0.262	0.227	0.243
InternVL3.5-30B-A3B2025.12		0.245	0.212	0.228
GPT-4o2025.12		0.221	0.333	0.28
LLaVA-OneVision-7B2025.12		0.211	0.194	0.202
GPT-5-nano2025.12		0.192	0.333	0.265