Visual Question Answering On Mm Vet

評価指標

GPT-4 score

評価結果

このベンチマークにおける各モデルのパフォーマンス結果

		Paper Title
gemini-2.0-flash-exp	81.2±0.4	-
gemini-exp-1206	78.1±0.2	-
Gemini 1.5 Pro (gemini-1.5-pro-002)	76.9±0.1	Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context
MMCTAgent (GPT-4 + GPT-4V)	74.24	MMCTAgent: Multi-modal Critical Thinking Agent Framework for Complex Visual Reasoning
Claude 3.5 Sonnet (claude-3-5-sonnet-20240620)	74.2±0.2	Claude 3.5 Sonnet Model Card Addendum
Qwen2-VL-72B	74.0	Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution
InternVL2.5-78B	72.3	Expanding Performance Boundaries of Open-Source Multimodal Models with Model, Data, and Test-Time Scaling
GPT-4o +text rationale +IoT	72.2	Image-of-Thought Prompting for Visual Reasoning Refinement in Multimodal Large Language Models
Lyra-Pro	71.4	Lyra: An Efficient and Speech-Centric Framework for Omni-Cognition
GLM-4V-Plus	71.1	CogVLM2: Visual Language Models for Image and Video Understanding
Phantom-7B	70.8	Phantom of Latent for Large Language and Vision Models
GPT-4o (gpt-4o-2024-05-13)	69.3±0.1	GPT-4 Technical Report
InternVL2.5-38B	68.8	Expanding Performance Boundaries of Open-Source Multimodal Models with Model, Data, and Test-Time Scaling
gpt-4o-mini-2024-07-18	68.6±0.1	GPT-4 Technical Report
GPT-4V	67.7±0.3	GPT-4 Technical Report
GPT-4V-Turbo-detail:high	67.6±0.1	GPT-4 Technical Report
Qwen-VL-Max	66.6±0.5	Qwen-VL: A Versatile Vision-Language Model for Understanding, Localization, Text Reading, and Beyond
Gemini 1.5 Pro (gemini-1.5-pro)	65.8±0.1	Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context
InternVL2-26B (SGP, token ratio 64%)	65.60	A Stitch in Time Saves Nine: Small VLM is a Precise Guidance for Accelerating Large VLMs
Baichuan-Omni (7B)	65.4	Baichuan-Omni Technical Report

0 of 229 row(s) selected.

Command Palette

Visual Question Answering On Mm Vet

評価指標

評価結果