AI Paper Atlas

Research map / Benchmarks / MMBench

MMBench leaderboard

169 papers at ICML, NeurIPS, ICLR, CVPR, AAAI, ICCV and ECCV, and the journals TPAMI, IJCV and JMLR report results on MMBench. Best accuracy: 92.65 (DREAM-R, ICML 2026).

Numbers are as each paper reports them, read by AI from its results tables and not checked by hand. Papers can differ in setting (training data, model size, extra data); the topic column shows what each paper is about. Open the paper before you compare or quote a number.

MMBench · accuracy

105 papers, each with its best reported number for its own method; higher is better.

#accuracy ↑MethodPaperTopic
192.65DREAM-R
Target Qwen3-VL-32B, draft Qwen3-VL-2B
DREAM-R: Multimodal Speculative Reasoning with RL-Based Refined Drafting, Precise Verification, and Fully Parallel Execution
ICML 2026 · Table 2
Reasoning and chain of thought
290.2C: LoRA-PID
Qwen2.5-VL-7B
Towards Understanding Modality Interaction in Multimodal Language Models via Partial Information Decomposition
ICML 2026 · Table 3
Cross-modal learning
389.5Ours (lookback)
Qwen3-VL-4B-Thinking
When to Think and When to Look: Uncertainty-Guided Lookback
CVPR 2026 · Table 2
Multimodal LLMs
489.06GPT-4o + VLB V2
GPT-4o
Dynamic Multimodal Evaluation with Flexible Complexity by Vision-Language Bootstrapping
ICLR 2025 · Table: image bootstrapping res
Multimodal LLMs
588.8L2-VMAS
Qwen3-VL-8B-Thinking
Dual Latent Memory for Visual Multi-agent System
ICML 2026 · Table 1
LLM agents
687.2+ DiG
Qwen3-VL-8B
DiG: Differential Grounding for Enhancing Fine-Grained Perception in Multimodal Large Language Models
CVPR 2026 · Table 2
Multimodal LLMs
786.4LLaVA-Prime-72B
Qwen2-72B
Learning Beyond Still Frames: Scaling Vision-Language Models with Video
ICCV 2025 · Table 7
Video-language understanding
885.4IPK (Ours)
LLaMA-7B + ViT/14
Leveraging Image as Compressed Visual Prompt and Hierarchical Visual Knowledge for Effective Image Utilization in MLLMs
AAAI 2026 · Table 1
Multimodal LLMs
985.4SophiaVL-R1-7B
Qwen2.5-VL-7B-Instruct
SophiaVL-R1: Reinforcing MLLMs Reasoning with Thinking Reward
ICLR 2026 · Table 2
Reasoning and chain of thought
1085.4Bi-VRL
LLaDA-V
Reinforcement Learning for Multimodal Diffusion Language Models via Bidimensional Trajectory and Thought Optimization
ECCV 2026 · Table 1
Multimodal LLMs
95 more papers on this leaderboard, the baselines reported on it and a CSV download are free with a scime account.
Sign in free to see all

MMBench · score

65 papers, each with its best reported number for its own method; higher is better.

#score ↑MethodPaperTopic
191.2UniDFlow (4B)Best of Both Worlds: Multimodal Reasoning and Generation via Unified Discrete Flow Matching
ICML 2026 · Table 1
Multimodal LLMs
286.3Oryx-1.5-32B
Qwen-2.5-32B
Oryx MLLM: On-Demand Spatial-Temporal Understanding at Arbitrary Resolution
ICLR 2025 · Image Understanding table
Video-language understanding
385.6Uni-CoTUni-CoT: Towards Unified Chain-of-Thought Reasoning Across Text and Vision
ICLR 2026 · Table 3
Multimodal LLMs
483.9DenseMLLM
DenseMLLM-4B
DenseMLLM: Standard Multimodal LLMs for Dense Prediction
ICML 2026 · Table 2
Multimodal LLMs
583.3LLaDA-V + MPS + MRS
LLaDA-V
Mitigating Mask Prior Drift and Positional Attention Collapse in Large Diffusion Vision-Language Models
ICML 2026 · Table 1
Multimodal LLMs
683EMOVA-7BEMOVA: Empowering Language Models to See, Hear and Speak with Vivid Emotions
CVPR 2025 · Table 2
Language, safety and interpretability
782.8Qwen2-VL-7B+Critic-V
Qwen2-VL-7B
Critic-V: VLM Critics Help Catch VLM Errors in Multimodal Reasoning
CVPR 2025 · Main results table
Multimodal LLMs
882.7BlueLM-V-3B
BlueLM-3B
BlueLM-V-3B: Algorithm and System Co-Design for Multimodal Large Language Models on Mobile Devices
CVPR 2025 · Table 3
Quantization
982.3Insight-V
Our Base Model
Insight-V: Exploring Long-Chain Visual Reasoning with Multimodal Large Language Models
CVPR 2025 · Table 1
Multimodal LLMs
1081.5ForeSight (7B)
Qwen2.5-VL-7B
See Further, Think Deeper: Advancing VLM's Reasoning Ability with Low-level Visual Cues and Reflection
CVPR 2026 · Table 6
Multimodal LLMs
55 more papers on this leaderboard, the baselines reported on it and a CSV download are free with a scime account.
Sign in free to see all

Not affiliated with any conference, journal or the benchmark's authors. Corrections: contact@latenticlabs.com.