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

Multimodal Search on MMSearch

74.27Accuracy

Gemini-3-Pro

Updated 25d ago

Evaluation Results

Method	Links
Gemini-3-Pro 2026.06		74.27
Qwen3-VL-30B + ODE-RL 2026.05		69.7
MM-DeepResearch 32B 2026.03		69
Gemini-3-Flash 2026.04		69
MM-DeepResearch-8B 2026.03		67.8
SenseNova-MARS-8B 2026.03		67.4
TAPO 2026.06		67.21
Claude-4-Sonnet 2026.05		67.2
SenseNova-MARS 2026.06		66.67
SkyWork-R1V4 2026.06		66.1
GPT-5.2 2026.06		66.08
Qwen3-VL-8B + ODE-RL 2026.05		66
Qwen3-VL-30B + ODE-SFT 2026.05		65.7
Qwen3-VL-8B-Instruct+SAPO 2026.06		64.33
DeepEyes-v2-7B 2026.03		63.7
Deepeyes2 2026.04		63.7
Deepeyes2 2026.04		63.7
Claude-3.7-Sonnet 2026.05		63.7
DeepEyesV2 2026.06		63.7
GPT-5 2026.05		62.7
MIA 2026.04		62.6
Gemini-3-Pro 2026.06		62.57
Gemini-3-Flash 2026.06		62.57
Qwen3-VL-8B + ODE-SFT 2026.05		61.7
MM-DeepResearch-7B 2026.03		61.4
ExpeL 2026.04		61.4
Memento 2026.04		61.4
DR-MMSearchAgent 2026.04		61
Unsupervised MIA 2026.04		60.2
Gemini-2.5-Flash 2026.06		59.06
Gemini-3-Pro 2026.03		59
Gemini-3-Flash 2026.06		57.31
Gemini-3-Flash 2026.03		57.3
ReasoningBank 2026.04		57.3
Qwen3-VL-30B 2026.05		57
Gemini-2.5 Pro 2026.05		55.7
No Memory 2026.04		55.6
WebWatcher-32B 2026.03		55.3
WebWatcher 2026.04		55.3
WebWatcher 2026.04		55.3
WebWatcher-32B 2026.05		55.3
RAG 2026.04		54.4
MMSearch-R1-7B 2026.03		53.8
MMSearch-R1 2026.04		53.8
MMSearch-R1-7B 2026.05		53.8
MMSearch-R1 2026.06		53.8
Gemini-2.5 Flash 2026.05		53.3
GPT-5 2026.06		52.63
GPT-5 2026.03		52.6
Qwen3-VL-Plus 2026.03		50.7
Qwen2.5-VL-32B-Instruct 2026.06		49.71
Qwen3-VL-30B 2026.05		49.3
Qwen3-VL-32B-Instruct 2026.06		49.12
GPT-4o 2026.06		49.12
GPT-4o 2026.03		49.1
WebWatcher-7B 2026.03		49.1
WebWatcher-7B 2026.05		49.1
Qwen3-VL-8B 2026.05		48.7
Qwen3-VL-8B-Instruct 2026.06		47.37
Qwen3-VL-8B 2026.03		47.3
Qwen3-VL-8B 2026.05		47.3
VSearcher 2026.03		47.2
GPT-5 2026.03		45.2
Qwen3-VL-32B 2026.03		44.4
GPT-5.4 2026.04		44.4
MMSearch-R1 2026.04		43.9
Qwen3-VL-30B-A3B-Thinking 2026.03		43.7
Mem0 2026.04		43.3
GPT-5.2 2026.06		43.27
Qwen2.5-VL-72B-Instruct 2026.03		41.5
DeepEyesV2 2026.03		40.9
A-Mem 2026.04		40.9
GPT-5 2026.05		40.7
Gemini-2.5 Pro 2026.05		39.8
Qwen3-VL-8B-Thinking 2026.03		39.1
GPT-4o-mini 2026.06		38.6
MMSearch-R1 2026.03		37.5
Qwen3-VL-8B 2026.03		37.4
GPT-5 2026.03		35.1
GPT-5 2026.06		35.09
Qwen3-VL-4B-Thinking 2026.03		35
Visual-ARFT-7B 2026.03		34.5
Visual-ARFT 2026.06		34.5
InternVL3.5-38B 2026.03		33.3
Qwen2.5-VL-7B-Instruct 2026.06		32.16
Gemini-2.5 Flash 2026.05		30.4
InternVL3.5-8B 2026.03		29.2
Qwen2.5-VL-32B+ReACT 2026.04		27.3
Qwen2.5-VL-32B 2026.04		27.3
Gemini-2.5-Pro 2026.04		26.9
Gemini2.5 Pro 2026.04		26.9
GPT-4o 2026.06		23.39
PIXELRAG 2026.06		22.7
GPT-4o 2026.04		22.2
GPT-4o 2026.04		22.2
Gemini-2.5-Flash 2026.06		21.64
Qwen2.5-VL-7B+ReACT 2026.04		21.1
Qwen2.5-VL-7B 2026.04		21.1
Claude-3.7-Sonnet 2026.05		21.1
PIXELRAG (base) 2026.06		20.3

Showing 100 of 119 rows