MathVista
The official leaderboard hasn’t scored a model released after Llama 4 Maverick (April 2025). Labs report their own numbers for 47 newer models, such as Seed2.1 Pro at 90.7%.
Math reasoning in visual contexts (charts, geometry, puzzles) from 31 datasets. Answer accuracy.
Near-saturated.
- Released
- October 2023
- Announced on X
- 11K views, 94 likes
- Kind
- Visual math and diagrams
- Level
- School
- Scored
- Checked automatically
- Answers
- Short answer
- Tasks
- All public
- Measures
- Capability
- Subject
- Mathematics
- Language
- English
- Interaction
- Single turn
- Refresh
- One fixed set
- Run by
- Anyone
- In views
- Vision
- Strongest model tested
- Kimi K2.5 #76 on the capability index
- Last in a launch post
- October 2026
- Last model added
- October 2026
- Scores refreshed
- November 18th, 2026
In the authors’ words
Large Language Models (LLMs) and Large Multimodal Models (LMMs) exhibit impressive problem-solving skills in many tasks and domains, but their ability in mathematical reasoning in visual contexts has not been systematically studied. To bridge this gap, we present MathVista, a benchmark designed to combine challenges from diverse mathematical and visual tasks. It consists of 6,141 examples, derived from 28 existing multimodal datasets involving mathematics and 3 newly created datasets (i.e., IQTest, FunctionQA, and PaperQA). Completing these tasks requires fine-grained, deep visual understanding and compositional reasoning, which all state-of-the-art foundation models find challenging. With MathVista, we have conducted a comprehensive, quantitative evaluation of 12 prominent foundation models. The best-performing GPT-4V model achieves an overall accuracy of 49.9%, substantially outperforming Bard, the second-best performer, by 15.1%. Our in-depth analysis reveals that the superiority of GPT-4V is mainly attributed to its enhanced visual perception and mathematical reasoning. However, GPT-4V still falls short of human performance by 10.4%, as it often struggles to understand complex figures and perform rigorous reasoning. This significant gap underscores the critical role that MathVista will play in the development of general-purpose AI agents capable of tackling mathematically intensive and visually rich real-world tasks. We further explore the new ability of self-verification, the application of self-consistency, and the interactive chatbot capabilities of GPT-4V, highlighting its promising potential for future research. The project is available at https://mathvista.github.io/.
Seed2.1 Pro leads with 90.7%
Best score per model as labs report them in their own cards, drawn striped. Labs pick their own settings, so these compare less cleanly than independent runs.
Doubao-1.5-pro leads with 79.5% across all independent runs
Best score per model across the official leaderboard, Epoch AI’s own runs and any board Epoch AI copied. Dotted bars are reruns by Epoch AI or another board, and each row names its source.
Seed2.1 Pro leads with 90.7%, claims included
Best score per model from every source, including labs’ own numbers and claims Epoch AI collected from reports and papers. Labs pick their own settings, so a claim can beat an independent run of the same model. Striped bars are claims, and dotted bars are reruns by Epoch AI or another board.
- 90.7
Seed2.1 Pro
- 90.3
Qwen3.5 397B-A17B
- 90.1 Kimi K2.5
- 89.8
Seed2.0 Pro
- 88.0
MiniCPM-V-4.7-35B-A3B
- 87.8
Qwen3.5-27B
- 87.7
Seed1.8
- 87.4
Qwen3.5-122B-A10B
- 87.4
Qwen3.6 27B
- 86.8 o3
- 86.4
Qwen 3.6 35B-A3B
- 86.2
Qwen3.5-35B-A3B
- 86.1
Qwen3.5-Omni-Plus
- 85.8
Qwen3-VL-235B-A22B
- 85.7
Qwen3.5-9B
- 85.6
Seed1.5-VL
- 85.2 GLM-4.6V
- 85.1
Qwen3.5-4B
- 85.0
EXAONE 4.5 33B
- 84.8
ERNIE 5.0
- 84.6 GLM-4.5V
- 84.3 o4-mini
- 84.0
STEP3-VL-10B
- 82.9
Keye-VL-2.0-30B-A3B
- 82.8
Nemotron 3 Nano Omni 30B-A3B
- 82.7
InternVL3.5-241B-A28B
- 82.5
ERNIE-4.5-VL-28B-A3B-Thinking
- 81.5
Intern-S1
- 80.7 GLM-4.1V-9B-Thinking
- 80.6
Command A+
- 80.1 Kimi-VL-A3B-Thinking-2506
- 77.9
LongCat-Flash-Omni
- 76.9
Nemotron Nano 12B v2 VL
- 76.7
Qwen3.5-2B
- 75.9
Qwen3-Omni-30B-A3B
- 75.2
Phi-4-Reasoning-Vision-15B
- 74.9 Kimi k1.5
- 74.8
Qwen2.5-VL-72B
- 73.7
Llama 4 Maverick
- 73.5
Command A Vision
- 73.1 GPT-4.1 mini
- 73.1
BAGEL-7B-MoT
- 72.2 GPT-4.1
- 71.3 Kimi-VL-A3B-Thinking
- 71.0 o1
- 70.7 Claude 3.5 Sonnet (October 2024)
- 70.7
Llama 4 Scout
- 70.1
Magistral Medium 1.0
- 69.0 Grok-2
- 68.9
Mistral Small 3.1
- 68.5 LFM2.5-VL-3B
- 67.9
Qwen2.5-Omni-7B
- 67.7 Claude 3.5 Sonnet
- 67.6
Gemma 3 27B
- 67.1
Mistral Small 3.2
- 66.8
Magistral Small 1.0
- 63.9
Gemini 1.5 Pro
- 63.8 GPT-4o
- 62.9
Gemma 3 12B
- 62.4
Phi-4-multimodal
- 62.2
Qwen3.5-0.8B
- 56.7 GPT-4o mini
- 56.2 GPT-4.1 nano
- 53.0
Gemini 1.0 Ultra
- 52.1
Gemini 1.5 Pro
- 51.7 LFM2-VL-1.6B
- 50.5 Claude 3 Opus
- 50.0
Gemma 3 4B
- 47.9 Claude 3 Sonnet
- 46.4 Claude 3 Haiku
- 43.9 Phi-3.5-vision-instruct
- 1
Seed2.1 Pro 90.7%
- 2
Qwen3.5 397B-A17B 90.3%
- 3 Kimi K2.5 90.1%
- 4
Seed2.0 Pro 89.8%
- 5
MiniCPM-V-4.7-35B-A3B 88%
- 6
Qwen3.5-27B 87.8%
- 7
Seed1.8 87.7%
- 8
Qwen3.5-122B-A10B 87.4%
- 9
Qwen3.6 27B 87.4%
- 10 o3 86.8%
- 11
Qwen 3.6 35B-A3B 86.4%
- 12
Qwen3.5-35B-A3B 86.2%
- 13
Qwen3.5-Omni-Plus 86.1%
- 14
Qwen3-VL-235B-A22B 85.8%
- 15
Qwen3.5-9B 85.7%
- 16
Seed1.5-VL 85.6%
- 17 GLM-4.6V 85.2%
- 18
Qwen3.5-4B 85.1%
- 19
EXAONE 4.5 33B 85%
- 20
ERNIE 5.0 84.8%
- 21 GLM-4.5V 84.6%
- 22 o4-mini 84.3%
- 23
STEP3-VL-10B 84%
- 24
Keye-VL-2.0-30B-A3B 82.9%
- 25
Nemotron 3 Nano Omni 30B-A3B 82.8%
- 26
InternVL3.5-241B-A28B 82.7%
- 27
ERNIE-4.5-VL-28B-A3B-Thinking 82.5%
- 28
Intern-S1 81.5%
- 29 GLM-4.1V-9B-Thinking 80.7%
- 30
Command A+ 80.6%
- 31 Kimi-VL-A3B-Thinking-2506 80.1%
- 32
LongCat-Flash-Omni 77.9%
- 33
Nemotron Nano 12B v2 VL 76.9%
- 34
Qwen3.5-2B 76.7%
- 35
Qwen3-Omni-30B-A3B 75.9%
- 36
Phi-4-Reasoning-Vision-15B 75.2%
- 37 Kimi k1.5 74.9%
- 38
Qwen2.5-VL-72B 74.8%
- 39
Llama 4 Maverick 73.7%
- 40
Command A Vision 73.5%
- 41 GPT-4.1 mini 73.1%
- 42
BAGEL-7B-MoT 73.1%
- 43 GPT-4.1 72.2%
- 44 Kimi-VL-A3B-Thinking 71.3%
- 45 o1 71%
- 46 Claude 3.5 Sonnet (October 2024) 70.7%
- 47
Llama 4 Scout 70.7%
- 48
Magistral Medium 1.0 70.1%
- 49 Grok-2 Aug 2024 69%
- 50
Mistral Small 3.1 68.9%
- 51 LFM2.5-VL-3B 68.5%
- 52
Qwen2.5-Omni-7B 67.9%
- 53 Claude 3.5 Sonnet 67.7%
- 54
Gemma 3 27B 67.6%
- 55
Mistral Small 3.2 67.1%
- 56
Magistral Small 1.0 66.8%
- 57
Gemini 1.5 Pro May 2024 63.9%
- 58 GPT-4o May 2024 63.8%
- 59
Gemma 3 12B 62.9%
- 60
Phi-4-multimodal 62.4%
- 61
Qwen3.5-0.8B 62.2%
- 62 GPT-4o mini 56.7%
- 63 GPT-4.1 nano 56.2%
- 64
Gemini 1.0 Ultra 53%
- 65
Gemini 1.5 Pro Feb 2024 52.1%
- 66 LFM2-VL-1.6B 51.7%
- 67 Claude 3 Opus 50.5%
- 68
Gemma 3 4B 50%
- 69 Claude 3 Sonnet 47.9%
- 70 Claude 3 Haiku 46.4%
- 71 Phi-3.5-vision-instruct 43.9%
Source: Labs’ model cards
- 79.5
Doubao-1.5-pro
- 77.1
Ovis2-34B
- 74.9 Kimi k1.5
- 73.9 o1
- 73.7
Llama 4 Maverick
- 73.1
Gemini 2.0 Flash
- 71.4
QVQ-72B-Preview
- 70.5 Qwen2-VL-72B-Instruct
- 69.4
Pixtral Large
- 69.0 Grok-2
- 68.1 Grok-2 mini
- 67.7 Claude 3.5 Sonnet
- 67.3
InternVL2-8B-MPO
- 67.2
Ovis1.6-Gemma2-9B
- 66.8
InternVL2-Pro
- 63.9
Gemini 1.5 Pro
- 63.8 GPT-4o
- 62.4
Phi-4-multimodal
- 59.9
InternVL-Chat-V1-2-Plus
- 58.4
Gemini 1.5 Flash
- 58.1 GPT-4 Turbo
- 58.0
Pixtral 12B
- 57.6
InternLM-XComposer2-VL-7B
- 53.0
Gemini 1.0 Ultra
- 52.1
Gemini 1.5 Pro
- 50.5 Claude 3 Opus
- 49.9 GPT-4V
- 47.9 Claude 3 Sonnet
- 47.7
InternVL-Chat-V1-2
- 46.5
LLaVA-v1.6-34B
- 46.4 Claude 3 Haiku
- 45.2
Gemini 1.0 Pro
- 44.5
Phi-3-vision-128k-instruct
- 43.9 Phi-3.5-vision-instruct
- 43.3
Qwen-VL-Plus
- 40.6
MiniCPM-V 2.0
- 35.9 MM1-7B-Chat
- 34.9
OmniLMM-12B
- 34.6 llava-v1.6-vicuna-7b
- 32.0 MM1-3B-Chat
- 27.6
LLaVA-v1.5-13B
- 25.3 instructblip-vicuna-7b
- 19.8
IDEFICS-9B-Instruct
- 1
Doubao-1.5-pro 79.5%
- 2
Ovis2-34B 77.1%
- 3 Kimi k1.5 74.9%
- 4 o1 73.9%
- 5
Llama 4 Maverick 73.7%
- 6
Gemini 2.0 Flash Feb 2025 73.1%
- 7
QVQ-72B-Preview 71.4%
- 8 Qwen2-VL-72B-Instruct 70.5%
- 9
Pixtral Large 69.4%
- 10 Grok-2 Aug 2024 69%
- 11 Grok-2 mini 68.1%
- 12 Claude 3.5 Sonnet 67.7%
- 13
InternVL2-8B-MPO 67.3%
- 14
Ovis1.6-Gemma2-9B 67.2%
- 15
InternVL2-Pro 66.8%
- 16
Gemini 1.5 Pro May 2024 63.9%
- 17 GPT-4o May 2024 63.8%
- 18
Phi-4-multimodal 62.4%
- 19
InternVL-Chat-V1-2-Plus 59.9%
- 20
Gemini 1.5 Flash May 2024 58.4%
- 21 GPT-4 Turbo Apr 2024 58.1%
- 22
Pixtral 12B 58%
- 23
InternLM-XComposer2-VL-7B 57.6%
- 24
Gemini 1.0 Ultra 53%
- 25
Gemini 1.5 Pro Feb 2024 52.1%
- 26 Claude 3 Opus 50.5%
- 27 GPT-4V 49.9%
- 28 Claude 3 Sonnet 47.9%
- 29
InternVL-Chat-V1-2 47.7%
- 30
LLaVA-v1.6-34B 46.5%
- 31 Claude 3 Haiku 46.4%
- 32
Gemini 1.0 Pro 45.2%
- 33
Phi-3-vision-128k-instruct 44.5%
- 34 Phi-3.5-vision-instruct 43.9%
- 35
Qwen-VL-Plus 43.3%
- 36
MiniCPM-V 2.0 40.6%
- 37 MM1-7B-Chat 35.9%
- 38
OmniLMM-12B 34.9%
- 39 llava-v1.6-vicuna-7b 34.6%
- 40 MM1-3B-Chat 32%
- 41
LLaVA-v1.5-13B 27.6%
- 42 instructblip-vicuna-7b 25.3%
- 43
IDEFICS-9B-Instruct 19.8%
Source: Official MathVista leaderboard
- 90.7
Seed2.1 Pro
- 90.3
Qwen3.5 397B-A17B
- 90.1 Kimi K2.5
- 89.8
Seed2.0 Pro
- 88.0
MiniCPM-V-4.7-35B-A3B
- 87.8
Qwen3.5-27B
- 87.7
Seed1.8
- 87.4
Qwen3.5-122B-A10B
- 87.4
Qwen3.6 27B
- 86.8 o3
- 86.4
Qwen 3.6 35B-A3B
- 86.2
Qwen3.5-35B-A3B
- 86.1
Qwen3.5-Omni-Plus
- 85.8
Qwen3-VL-235B-A22B
- 85.7
Qwen3.5-9B
- 85.6
Seed1.5-VL
- 85.2 GLM-4.6V
- 85.1
Qwen3.5-4B
- 85.0
EXAONE 4.5 33B
- 84.8
ERNIE 5.0
- 84.6 GLM-4.5V
- 84.3 o4-mini
- 84.0
STEP3-VL-10B
- 82.9
Keye-VL-2.0-30B-A3B
- 82.8
Nemotron 3 Nano Omni 30B-A3B
- 82.7
InternVL3.5-241B-A28B
- 82.5
ERNIE-4.5-VL-28B-A3B-Thinking
- 81.5
Intern-S1
- 80.7 GLM-4.1V-9B-Thinking
- 80.6
Command A+
- 80.1 Kimi-VL-A3B-Thinking-2506
- 79.5
Doubao-1.5-pro
- 77.9
LongCat-Flash-Omni
- 77.1
Ovis2-34B
- 76.9
Nemotron Nano 12B v2 VL
- 76.7
Qwen3.5-2B
- 75.9
Qwen3-Omni-30B-A3B
- 75.2
Phi-4-Reasoning-Vision-15B
- 74.9 Kimi k1.5
- 74.8
Qwen2.5-VL-72B
- 73.9 o1
- 73.7
Llama 4 Maverick
- 73.5
Command A Vision
- 73.1
Gemini 2.0 Flash
- 73.1 GPT-4.1 mini
- 73.1
BAGEL-7B-MoT
- 72.2 GPT-4.1
- 71.4
QVQ-72B-Preview
- 71.3 Kimi-VL-A3B-Thinking
- 70.7 Claude 3.5 Sonnet (October 2024)
- 70.7
Llama 4 Scout
- 70.5 Qwen2-VL-72B-Instruct
- 70.1
Magistral Medium 1.0
- 69.4
Pixtral Large
- 69.0 Grok-2
- 68.9
Mistral Small 3.1
- 68.5 LFM2.5-VL-3B
- 68.1 Grok-2 mini
- 67.9
Qwen2.5-Omni-7B
- 67.7 Claude 3.5 Sonnet
- 67.6
Gemma 3 27B
- 67.3
InternVL2-8B-MPO
- 67.2
Ovis1.6-Gemma2-9B
- 67.1
Mistral Small 3.2
- 66.8
InternVL2-Pro
- 66.8
Magistral Small 1.0
- 63.9
Gemini 1.5 Pro
- 63.8 GPT-4o
- 62.9
Gemma 3 12B
- 62.4
Phi-4-multimodal
- 62.2
Qwen3.5-0.8B
- 59.9
InternVL-Chat-V1-2-Plus
- 58.4
Gemini 1.5 Flash
- 58.1 GPT-4 Turbo
- 58.0
Pixtral 12B
- 57.6
InternLM-XComposer2-VL-7B
- 56.7 GPT-4o mini
- 56.2 GPT-4.1 nano
- 53.0
Gemini 1.0 Ultra
- 52.1
Gemini 1.5 Pro
- 51.7 LFM2-VL-1.6B
- 50.5 Claude 3 Opus
- 50.0
Gemma 3 4B
- 49.9 GPT-4V
- 47.9 Claude 3 Sonnet
- 47.7
InternVL-Chat-V1-2
- 46.5
LLaVA-v1.6-34B
- 46.4 Claude 3 Haiku
- 45.2
Gemini 1.0 Pro
- 44.5
Phi-3-vision-128k-instruct
- 43.9 Phi-3.5-vision-instruct
- 43.3
Qwen-VL-Plus
- 40.6
MiniCPM-V 2.0
- 35.9 MM1-7B-Chat
- 34.9
OmniLMM-12B
- 34.6 llava-v1.6-vicuna-7b
- 32.0 MM1-3B-Chat
- 27.6
LLaVA-v1.5-13B
- 25.3 instructblip-vicuna-7b
- 19.8
IDEFICS-9B-Instruct
- 1
Seed2.1 Pro 90.7%
- 2
Qwen3.5 397B-A17B 90.3%
- 3 Kimi K2.5 90.1%
- 4
Seed2.0 Pro 89.8%
- 5
MiniCPM-V-4.7-35B-A3B 88%
- 6
Qwen3.5-27B 87.8%
- 7
Seed1.8 87.7%
- 8
Qwen3.5-122B-A10B 87.4%
- 9
Qwen3.6 27B 87.4%
- 10 o3 86.8%
- 11
Qwen 3.6 35B-A3B 86.4%
- 12
Qwen3.5-35B-A3B 86.2%
- 13
Qwen3.5-Omni-Plus 86.1%
- 14
Qwen3-VL-235B-A22B 85.8%
- 15
Qwen3.5-9B 85.7%
- 16
Seed1.5-VL 85.6%
- 17 GLM-4.6V 85.2%
- 18
Qwen3.5-4B 85.1%
- 19
EXAONE 4.5 33B 85%
- 20
ERNIE 5.0 84.8%
- 21 GLM-4.5V 84.6%
- 22 o4-mini 84.3%
- 23
STEP3-VL-10B 84%
- 24
Keye-VL-2.0-30B-A3B 82.9%
- 25
Nemotron 3 Nano Omni 30B-A3B 82.8%
- 26
InternVL3.5-241B-A28B 82.7%
- 27
ERNIE-4.5-VL-28B-A3B-Thinking 82.5%
- 28
Intern-S1 81.5%
- 29 GLM-4.1V-9B-Thinking 80.7%
- 30
Command A+ 80.6%
- 31 Kimi-VL-A3B-Thinking-2506 80.1%
- 32
Doubao-1.5-pro 79.5%
- 33
LongCat-Flash-Omni 77.9%
- 34
Ovis2-34B 77.1%
- 35
Nemotron Nano 12B v2 VL 76.9%
- 36
Qwen3.5-2B 76.7%
- 37
Qwen3-Omni-30B-A3B 75.9%
- 38
Phi-4-Reasoning-Vision-15B 75.2%
- 39 Kimi k1.5 74.9%
- 40
Qwen2.5-VL-72B 74.8%
- 41 o1 73.9%
- 42
Llama 4 Maverick 73.7%
- 43
Command A Vision 73.5%
- 44
Gemini 2.0 Flash Feb 2025 73.1%
- 45 GPT-4.1 mini 73.1%
- 46
BAGEL-7B-MoT 73.1%
- 47 GPT-4.1 72.2%
- 48
QVQ-72B-Preview 71.4%
- 49 Kimi-VL-A3B-Thinking 71.3%
- 50 Claude 3.5 Sonnet (October 2024) 70.7%
- 51
Llama 4 Scout 70.7%
- 52 Qwen2-VL-72B-Instruct 70.5%
- 53
Magistral Medium 1.0 70.1%
- 54
Pixtral Large 69.4%
- 55 Grok-2 Aug 2024 69%
- 56
Mistral Small 3.1 68.9%
- 57 LFM2.5-VL-3B 68.5%
- 58 Grok-2 mini 68.1%
- 59
Qwen2.5-Omni-7B 67.9%
- 60 Claude 3.5 Sonnet 67.7%
- 61
Gemma 3 27B 67.6%
- 62
InternVL2-8B-MPO 67.3%
- 63
Ovis1.6-Gemma2-9B 67.2%
- 64
Mistral Small 3.2 67.1%
- 65
InternVL2-Pro 66.8%
- 66
Magistral Small 1.0 66.8%
- 67
Gemini 1.5 Pro May 2024 63.9%
- 68 GPT-4o May 2024 63.8%
- 69
Gemma 3 12B 62.9%
- 70
Phi-4-multimodal 62.4%
- 71
Qwen3.5-0.8B 62.2%
- 72
InternVL-Chat-V1-2-Plus 59.9%
- 73
Gemini 1.5 Flash May 2024 58.4%
- 74 GPT-4 Turbo Apr 2024 58.1%
- 75
Pixtral 12B 58%
- 76
InternLM-XComposer2-VL-7B 57.6%
- 77 GPT-4o mini 56.7%
- 78 GPT-4.1 nano 56.2%
- 79
Gemini 1.0 Ultra 53%
- 80
Gemini 1.5 Pro Feb 2024 52.1%
- 81 LFM2-VL-1.6B 51.7%
- 82 Claude 3 Opus 50.5%
- 83
Gemma 3 4B 50%
- 84 GPT-4V 49.9%
- 85 Claude 3 Sonnet 47.9%
- 86
InternVL-Chat-V1-2 47.7%
- 87
LLaVA-v1.6-34B 46.5%
- 88 Claude 3 Haiku 46.4%
- 89
Gemini 1.0 Pro 45.2%
- 90
Phi-3-vision-128k-instruct 44.5%
- 91 Phi-3.5-vision-instruct 43.9%
- 92
Qwen-VL-Plus 43.3%
- 93
MiniCPM-V 2.0 40.6%
- 94 MM1-7B-Chat 35.9%
- 95
OmniLMM-12B 34.9%
- 96 llava-v1.6-vicuna-7b 34.6%
- 97 MM1-3B-Chat 32%
- 98
LLaVA-v1.5-13B 27.6%
- 99 instructblip-vicuna-7b 25.3%
- 100
IDEFICS-9B-Instruct 19.8%
Sources: Labs’ model cards and official MathVista leaderboard
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | | 90.7% | Jun 2026 | Lab model card | |
| 2 | | 90.3% | Feb 2026 | Lab model card | |
| 3 | Kimi K2.5 | 90.1% | Jan 2026 | Lab model card | |
| 4 | | 89.8% | Feb 2026 | Lab model card | |
| 5 | | 88% | Oct 2026 | Lab model card | |
| 6 | | 87.8% | Feb 2026 | Lab model card | |
| 7 | | 87.7% | Dec 2025 | Lab model card | |
| 8 | | 87.4% | Feb 2026 | Lab model card | |
| 9 | | 87.4% | Apr 2026 | Lab model card | |
| 10 | o3 | 86.8% | Apr 2025 | Lab model card | |
| 11 | | 86.4% | Apr 2026 | Lab model card | |
| 12 | | 86.2% | Feb 2026 | Lab model card | |
| 13 | | 86.1% | Mar 2026 | Lab model card | |
| 14 | | 85.8% | Sep 2025 | Lab model card | |
| 15 | | 85.7% | Mar 2026 | Lab model card | |
| 16 | | 85.6% | May 2025 | Lab model card | |
| 17 | GLM-4.6V | 85.2% | Dec 2025 | Lab model card | |
| 18 | | 85.1% | Mar 2026 | Lab model card | |
| 19 | | 85% | Apr 2026 | Lab model card | |
| 20 | | 84.8% | Feb 2026 | Lab model card | |
| 21 | GLM-4.5V | 84.6% | Aug 2025 | Lab model card | |
| 22 | o4-mini | 84.3% | Apr 2025 | Lab model card | |
| 23 | | 84% | Jan 2026 | Lab model card | |
| 24 | | 82.9% | May 2026 | Lab model card | |
| 25 | | 82.8% | Apr 2026 | Lab model card | |
| 26 | | 82.7% | Aug 2025 | Lab model card | |
| 27 | | 82.5% | Nov 2025 | Lab model card | |
| 28 | | 81.5% | Jul 2025 | Lab model card | |
| 29 | GLM-4.1V-9B-Thinking | 80.7% | Jul 2025 | Lab model card | |
| 30 | | 80.6% | May 2026 | Lab model card | |
| 31 | Kimi-VL-A3B-Thinking-2506 | 80.1% | Jun 2025 | Lab model card | |
| 32 | | 77.9% | Oct 2025 | Lab model card | |
| 33 | | 76.9% | Oct 2025 | Lab model card | |
| 34 | | 76.7% | Mar 2026 | Lab model card | |
| 35 | | 75.9% | Sep 2025 | Lab model card | |
| 36 | | 75.2% | Mar 2026 | Lab model card | |
| 37 | Kimi k1.5 | 74.9% | Jan 2025 | Lab model card | |
| 38 | | 74.8% | Jan 2025 | Lab model card | |
| 39 | | 73.7% | Apr 2025 | Lab model card | |
| 40 | | 73.5% | Jul 2025 | Lab model card | |
| 41 | GPT-4.1 mini | 73.1% | Apr 2025 | Lab model card | |
| 42 | | 73.1% | May 2025 | Lab model card | |
| 43 | GPT-4.1 | 72.2% | Apr 2025 | Lab model card | |
| 44 | Kimi-VL-A3B-Thinking | 71.3% | Apr 2025 | Lab model card | |
| 45 | o1 | 71% | Dec 2024 | Lab model card | |
| 46 | Claude 3.5 Sonnet (October 2024) | 70.7% | Oct 2024 | Lab model card | |
| 47 | | 70.7% | Apr 2025 | Lab model card | |
| 48 | | 70.1% | Jun 2025 | Lab model card | |
| 49 | Grok-2 Aug 2024 | 69% | Aug 2024 | Lab model card | |
| 50 | | 68.9% | Mar 2025 | Lab model card | |
| 51 | LFM2.5-VL-3B | 68.5% | Aug 2026 | Lab model card | |
| 52 | | 67.9% | Mar 2025 | Lab model card | |
| 53 | Claude 3.5 Sonnet | 67.7% | Jun 2024 | Lab model card | |
| 54 | | 67.6% | Mar 2025 | Lab model card | |
| 55 | | 67.1% | Jun 2025 | Lab model card | |
| 56 | | 66.8% | Jun 2025 | Lab model card | |
| 57 | | 63.9% | May 2024 | Lab model card | |
| 58 | GPT-4o May 2024 | 63.8% | May 2024 | Lab model card | |
| 59 | | 62.9% | Mar 2025 | Lab model card | |
| 60 | | 62.4% | Feb 2025 | Lab model card | |
| 61 | | 62.2% | Mar 2026 | Lab model card | |
| 62 | GPT-4o mini | 56.7% | Jul 2024 | Lab model card | |
| 63 | GPT-4.1 nano | 56.2% | Apr 2025 | Lab model card | |
| 64 | | 53% | Dec 2023 | Lab model card | |
| 65 | | 52.1% | Feb 2024 | Lab model card | |
| 66 | LFM2-VL-1.6B | 51.7% | Nov 2025 | Lab model card | |
| 67 | Claude 3 Opus | 50.5% | Mar 2024 | Lab model card | |
| 68 | | 50% | Mar 2025 | Lab model card | |
| 69 | Claude 3 Sonnet | 47.9% | Mar 2024 | Lab model card | |
| 70 | Claude 3 Haiku | 46.4% | Mar 2024 | Lab model card | |
| 71 | Phi-3.5-vision-instruct | 43.9% | Aug 2024 | Lab model card |
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | 79.5% | Jan 2025 | Official leaderboard | ||
| 2 | 77.1% | Feb 2025 | Official leaderboard | ||
| 3 | Kimi k1.5 | 74.9% | Jan 2025 | Official leaderboard | |
| 4 | o1 | 73.9% | Dec 2024 | Official leaderboard | |
| 5 | 73.7% | Apr 2025 | Official leaderboard | ||
| 6 | 73.1% | Feb 2025 | Official leaderboard | ||
| 7 | 71.4% | Dec 2024 | Official leaderboard | ||
| 8 | Qwen2-VL-72B-Instruct | 70.5% | Aug 2024 | Official leaderboard | |
| 9 | 69.4% | Nov 2024 | Official leaderboard | ||
| 10 | Grok-2 Aug 2024 | 69% | Aug 2024 | Official leaderboard | |
| 11 | Grok-2 mini | 68.1% | Aug 2024 | Official leaderboard | |
| 12 | Claude 3.5 Sonnet | 67.7% | Jun 2024 | Official leaderboard | |
| 13 | 67.3% | Nov 2024 | Official leaderboard | ||
| 14 | 67.2% | Sep 2024 | Official leaderboard | ||
| 15 | 66.8% | Jul 2024 | Official leaderboard | ||
| 16 | 63.9% | May 2024 | Official leaderboard | ||
| 17 | GPT-4o May 2024 | 63.8% | May 2024 | Official leaderboard | |
| 18 | 62.4% | Feb 2025 | Official leaderboard | ||
| 19 | 59.9% | Feb 2024 | Official leaderboard | ||
| 20 | 58.4% | May 2024 | Official leaderboard | ||
| 21 | GPT-4 Turbo Apr 2024 | 58.1% | Apr 2024 | Official leaderboard | |
| 22 | 58% | Sep 2024 | Official leaderboard | ||
| 23 | 57.6% | Jan 2024 | Official leaderboard | ||
| 24 | 53% | Dec 2023 | Official leaderboard | ||
| 25 | 52.1% | Feb 2024 | Official leaderboard | ||
| 26 | Claude 3 Opus | 50.5% | Feb 2024 | Official leaderboard | |
| 27 | GPT-4V | 49.9% | Nov 2023 | Official leaderboard | |
| 28 | Claude 3 Sonnet | 47.9% | Feb 2024 | Official leaderboard | |
| 29 | 47.7% | Feb 2024 | Official leaderboard | ||
| 30 | 46.5% | Jan 2024 | Official leaderboard | ||
| 31 | Claude 3 Haiku | 46.4% | Mar 2024 | Official leaderboard | |
| 32 | 45.2% | Dec 2023 | Official leaderboard | ||
| 33 | 44.5% | May 2024 | Official leaderboard | ||
| 34 | Phi-3.5-vision-instruct | 43.9% | Aug 2024 | Official leaderboard | |
| 35 | 43.3% | Jan 2024 | Official leaderboard | ||
| 36 | 40.6% | Apr 2024 | Official leaderboard | ||
| 37 | MM1-7B-Chat | 35.9% | Mar 2024 | Official leaderboard | |
| 38 | 34.9% | Jan 2024 | Official leaderboard | ||
| 39 | llava-v1.6-vicuna-7b | 34.6% | Jan 2024 | Official leaderboard | |
| 40 | MM1-3B-Chat | 32% | Mar 2024 | Official leaderboard | |
| 41 | 27.6% | Oct 2023 | Official leaderboard | ||
| 42 | instructblip-vicuna-7b | 25.3% | May 2023 | Official leaderboard | |
| 43 | 19.8% | Jul 2023 | Official leaderboard |
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | 90.7% | Jun 2026 | Lab model card | ||
| 2 | 90.3% | Feb 2026 | Lab model card | ||
| 3 | Kimi K2.5 | 90.1% | Jan 2026 | Lab model card | |
| 4 | 89.8% | Feb 2026 | Lab model card | ||
| 5 | 88% | Oct 2026 | Lab model card | ||
| 6 | 87.8% | Feb 2026 | Lab model card | ||
| 7 | 87.7% | Dec 2025 | Lab model card | ||
| 8 | 87.4% | Feb 2026 | Lab model card | ||
| 9 | 87.4% | Apr 2026 | Lab model card | ||
| 10 | o3 | 86.8% | Apr 2025 | Lab model card | |
| 11 | 86.4% | Apr 2026 | Lab model card | ||
| 12 | 86.2% | Feb 2026 | Lab model card | ||
| 13 | 86.1% | Mar 2026 | Lab model card | ||
| 14 | 85.8% | Sep 2025 | Lab model card | ||
| 15 | 85.7% | Mar 2026 | Lab model card | ||
| 16 | 85.6% | May 2025 | Lab model card | ||
| 17 | GLM-4.6V | 85.2% | Dec 2025 | Lab model card | |
| 18 | 85.1% | Mar 2026 | Lab model card | ||
| 19 | 85% | Apr 2026 | Lab model card | ||
| 20 | 84.8% | Feb 2026 | Lab model card | ||
| 21 | GLM-4.5V | 84.6% | Aug 2025 | Lab model card | |
| 22 | o4-mini | 84.3% | Apr 2025 | Lab model card | |
| 23 | 84% | Jan 2026 | Lab model card | ||
| 24 | 82.9% | May 2026 | Lab model card | ||
| 25 | 82.8% | Apr 2026 | Lab model card | ||
| 26 | 82.7% | Aug 2025 | Lab model card | ||
| 27 | 82.5% | Nov 2025 | Lab model card | ||
| 28 | 81.5% | Jul 2025 | Lab model card | ||
| 29 | GLM-4.1V-9B-Thinking | 80.7% | Jul 2025 | Lab model card | |
| 30 | 80.6% | May 2026 | Lab model card | ||
| 31 | Kimi-VL-A3B-Thinking-2506 | 80.1% | Jun 2025 | Lab model card | |
| 32 | 79.5% | Jan 2025 | Official leaderboard | ||
| 33 | 77.9% | Oct 2025 | Lab model card | ||
| 34 | 77.1% | Feb 2025 | Official leaderboard | ||
| 35 | 76.9% | Oct 2025 | Lab model card | ||
| 36 | 76.7% | Mar 2026 | Lab model card | ||
| 37 | 75.9% | Sep 2025 | Lab model card | ||
| 38 | 75.2% | Mar 2026 | Lab model card | ||
| 39 | Kimi k1.5 | 74.9% | Jan 2025 | Official leaderboard | |
| 40 | 74.8% | Jan 2025 | Lab model card | ||
| 41 | o1 | 73.9% | Dec 2024 | Official leaderboard | |
| 42 | 73.7% | Apr 2025 | Official leaderboard | ||
| 43 | 73.5% | Jul 2025 | Lab model card | ||
| 44 | 73.1% | Feb 2025 | Official leaderboard | ||
| 45 | GPT-4.1 mini | 73.1% | Apr 2025 | Lab model card | |
| 46 | 73.1% | May 2025 | Lab model card | ||
| 47 | GPT-4.1 | 72.2% | Apr 2025 | Lab model card | |
| 48 | 71.4% | Dec 2024 | Official leaderboard | ||
| 49 | Kimi-VL-A3B-Thinking | 71.3% | Apr 2025 | Lab model card | |
| 50 | Claude 3.5 Sonnet (October 2024) | 70.7% | Oct 2024 | Lab model card | |
| 51 | 70.7% | Apr 2025 | Lab model card | ||
| 52 | Qwen2-VL-72B-Instruct | 70.5% | Aug 2024 | Official leaderboard | |
| 53 | 70.1% | Jun 2025 | Lab model card | ||
| 54 | 69.4% | Nov 2024 | Official leaderboard | ||
| 55 | Grok-2 Aug 2024 | 69% | Aug 2024 | Official leaderboard | |
| 56 | 68.9% | Mar 2025 | Lab model card | ||
| 57 | LFM2.5-VL-3B | 68.5% | Aug 2026 | Lab model card | |
| 58 | Grok-2 mini | 68.1% | Aug 2024 | Official leaderboard | |
| 59 | 67.9% | Mar 2025 | Lab model card | ||
| 60 | Claude 3.5 Sonnet | 67.7% | Jun 2024 | Official leaderboard | |
| 61 | 67.6% | Mar 2025 | Lab model card | ||
| 62 | 67.3% | Nov 2024 | Official leaderboard | ||
| 63 | 67.2% | Sep 2024 | Official leaderboard | ||
| 64 | 67.1% | Jun 2025 | Lab model card | ||
| 65 | 66.8% | Jul 2024 | Official leaderboard | ||
| 66 | 66.8% | Jun 2025 | Lab model card | ||
| 67 | 63.9% | May 2024 | Official leaderboard | ||
| 68 | GPT-4o May 2024 | 63.8% | May 2024 | Official leaderboard | |
| 69 | 62.9% | Mar 2025 | Lab model card | ||
| 70 | 62.4% | Feb 2025 | Official leaderboard | ||
| 71 | 62.2% | Mar 2026 | Lab model card | ||
| 72 | 59.9% | Feb 2024 | Official leaderboard | ||
| 73 | 58.4% | May 2024 | Official leaderboard | ||
| 74 | GPT-4 Turbo Apr 2024 | 58.1% | Apr 2024 | Official leaderboard | |
| 75 | 58% | Sep 2024 | Official leaderboard | ||
| 76 | 57.6% | Jan 2024 | Official leaderboard | ||
| 77 | GPT-4o mini | 56.7% | Jul 2024 | Lab model card | |
| 78 | GPT-4.1 nano | 56.2% | Apr 2025 | Lab model card | |
| 79 | 53% | Dec 2023 | Official leaderboard | ||
| 80 | 52.1% | Feb 2024 | Official leaderboard | ||
| 81 | LFM2-VL-1.6B | 51.7% | Nov 2025 | Lab model card | |
| 82 | Claude 3 Opus | 50.5% | Feb 2024 | Official leaderboard | |
| 83 | 50% | Mar 2025 | Lab model card | ||
| 84 | GPT-4V | 49.9% | Nov 2023 | Official leaderboard | |
| 85 | Claude 3 Sonnet | 47.9% | Feb 2024 | Official leaderboard | |
| 86 | 47.7% | Feb 2024 | Official leaderboard | ||
| 87 | 46.5% | Jan 2024 | Official leaderboard | ||
| 88 | Claude 3 Haiku | 46.4% | Mar 2024 | Official leaderboard | |
| 89 | 45.2% | Dec 2023 | Official leaderboard | ||
| 90 | 44.5% | May 2024 | Official leaderboard | ||
| 91 | Phi-3.5-vision-instruct | 43.9% | Aug 2024 | Official leaderboard | |
| 92 | 43.3% | Jan 2024 | Official leaderboard | ||
| 93 | 40.6% | Apr 2024 | Official leaderboard | ||
| 94 | MM1-7B-Chat | 35.9% | Mar 2024 | Official leaderboard | |
| 95 | 34.9% | Jan 2024 | Official leaderboard | ||
| 96 | llava-v1.6-vicuna-7b | 34.6% | Jan 2024 | Official leaderboard | |
| 97 | MM1-3B-Chat | 32% | Mar 2024 | Official leaderboard | |
| 98 | 27.6% | Oct 2023 | Official leaderboard | ||
| 99 | instructblip-vicuna-7b | 25.3% | May 2023 | Official leaderboard | |
| 100 | 19.8% | Jul 2023 | Official leaderboard |
The benchmark's own leaderboard has 76 entries, 43 of them models we track. Switch the scores above to rank them together.
Rankings by part
We read 12 parts from the benchmark’s own leaderboard in 2 groups: task type and domain. Each ranks models on one part of the test. Doubao-1.5-pro leads 10 of them, and Ovis2-34B leads the other 2. Arrows show how far a model moved from its overall rank among the models we track.
Figure question answering
15 models we track have a score here.
- 1
Doubao-1.5-pro 77.7%
- 2
Ovis2-34B 75.5%
- 3
InternVL2-8B-MPO ↑10 72.5%
- 4
InternVL2-Pro ↑11 70.6%
- 5
InternLM-XComposer2-VL-7B ↑18 55%
- 6
Qwen-VL-Plus ↑29 54.6%
- 7
MiniCPM-V 2.0 ↑29 53.2%
- 8
InternVL-Chat-V1-2-Plus ↑11 51.7%
- 9
InternVL-Chat-V1-2 ↑20 50.9%
- 10
Gemini 1.0 Ultra ↑14 49.1%
Geometry problem solving
15 models we track have a score here.
- 1
Doubao-1.5-pro 88.9%
- 2
Ovis2-34B 84.6%
- 3
InternVL2-8B-MPO ↑10 73.6%
- 4
InternVL2-Pro ↑11 65.4%
- 5
InternLM-XComposer2-VL-7B ↑18 63%
- 6
InternVL-Chat-V1-2 ↑23 61.1%
- 7
InternVL-Chat-V1-2-Plus ↑12 61.1%
- 8
Gemini 1.0 Ultra ↑16 56.2%
- 9 GPT-4V ↑18 50.5%
- 10
Gemini 1.0 Pro ↑22 40.4%
Math word problem
15 models we track have a score here.
- 1
Ovis2-34B ↑1 89.2%
- 2
Doubao-1.5-pro ↓1 86%
- 3
InternVL-Chat-V1-2-Plus ↑16 79.6%
- 4
InternVL2-Pro ↑11 76.9%
- 5
InternLM-XComposer2-VL-7B ↑18 73.7%
- 6
InternVL2-8B-MPO ↑7 69.9%
- 7 GPT-4V ↑20 57.5%
- 8
Gemini 1.0 Ultra ↑16 53.8%
- 9
Gemini 1.0 Pro ↑23 39.2%
- 10
MiniCPM-V 2.0 ↑26 37.1%
Textbook question answering
15 models we track have a score here.
- 1
Doubao-1.5-pro 82.3%
- 2
Ovis2-34B 72.8%
- 3
InternVL2-Pro ↑12 71.5%
- 4
Gemini 1.0 Ultra ↑20 69%
- 5
InternVL2-8B-MPO ↑8 66.5%
- 6 GPT-4V ↑21 65.2%
- 7
Gemini 1.0 Pro ↑25 61.4%
- 8
InternLM-XComposer2-VL-7B ↑15 56.3%
- 9
Qwen-VL-Plus ↑26 55.1%
- 10
InternVL-Chat-V1-2-Plus ↑9 52.5%
Visual question answering
15 models we track have a score here.
- 1
Doubao-1.5-pro 62%
- 2
Ovis2-34B 62%
- 3
InternVL-Chat-V1-2-Plus ↑16 57%
- 4
InternVL2-8B-MPO ↑9 50.3%
- 5
InternVL2-Pro ↑10 48%
- 6
InternVL-Chat-V1-2 ↑23 44.7%
- 7
Gemini 1.0 Ultra ↑17 40.2%
- 8
InternLM-XComposer2-VL-7B ↑15 39.7%
- 9
Gemini 1.0 Pro ↑23 39.1%
- 10
OmniLMM-12B ↑28 39.1%
Algebraic
15 models we track have a score here.
- 1
Doubao-1.5-pro 89.7%
- 2
Ovis2-34B 80.1%
- 3
InternVL2-8B-MPO ↑10 70.1%
- 4
InternVL2-Pro ↑11 66.5%
- 5
Gemini 1.0 Ultra ↑19 58.4%
- 6
InternLM-XComposer2-VL-7B ↑17 56.6%
- 7
InternVL-Chat-V1-2-Plus ↑12 54.5%
- 8 GPT-4V ↑19 53%
- 9
InternVL-Chat-V1-2 ↑20 52.3%
- 10
Gemini 1.0 Pro ↑22 45.2%
Arithmetic
15 models we track have a score here.
- 1
Doubao-1.5-pro 73.5%
- 2
Ovis2-34B 72.2%
- 3
InternVL-Chat-V1-2-Plus ↑16 63.2%
- 4
InternVL2-Pro ↑11 62.3%
- 5
InternVL2-8B-MPO ↑8 57.5%
- 6
InternLM-XComposer2-VL-7B ↑17 52.4%
- 7 GPT-4V ↑20 49%
- 8
Gemini 1.0 Ultra ↑16 45.9%
- 9
Gemini 1.0 Pro ↑23 38.8%
- 10
InternVL-Chat-V1-2 ↑19 36.5%
Geometry
15 models we track have a score here.
- 1
Doubao-1.5-pro 88.3%
- 2
Ovis2-34B 82%
- 3
InternVL2-8B-MPO ↑10 71.5%
- 4
InternVL2-Pro ↑11 63.6%
- 5
InternLM-XComposer2-VL-7B ↑18 62.3%
- 6
InternVL-Chat-V1-2-Plus ↑13 61.1%
- 7
InternVL-Chat-V1-2 ↑22 58.2%
- 8
Gemini 1.0 Ultra ↑16 55.6%
- 9 GPT-4V ↑18 51%
- 10
Gemini 1.0 Pro ↑22 41%
Logical
15 models we track have a score here.
- 1
Ovis2-34B ↑1 37.8%
- 2
Doubao-1.5-pro ↓1 35.1%
- 3
InternVL2-Pro ↑12 27%
- 4
InternVL2-8B-MPO ↑9 27%
- 5 GPT-4V ↑22 21.6%
- 6
Gemini 1.0 Ultra ↑18 21.6%
- 7 instructblip-vicuna-7b ↑35 18.9%
- 8
IDEFICS-9B-Instruct ↑35 18.9%
- 9
Qwen-VL-Plus ↑26 18.9%
- 10
OmniLMM-12B ↑28 18.9%
Numeric
15 models we track have a score here.
- 1
Doubao-1.5-pro 55.3%
- 2
Ovis2-34B 54.9%
- 3
InternVL-Chat-V1-2-Plus ↑16 48.6%
- 4
InternVL2-8B-MPO ↑9 43.1%
- 5
InternLM-XComposer2-VL-7B ↑18 42.4%
- 6
InternVL2-Pro ↑9 40.3%
- 7
MiniCPM-V 2.0 ↑29 39.6%
- 8
Gemini 1.0 Ultra ↑16 38.9%
- 9
Gemini 1.0 Pro ↑23 32.6%
- 10
InternVL-Chat-V1-2 ↑19 30.6%
Scientific
15 models we track have a score here.
- 1
Doubao-1.5-pro 77.1%
- 2
Ovis2-34B 73.8%
- 3
InternVL2-Pro ↑12 65.6%
- 4
InternVL2-8B-MPO ↑9 65.6%
- 5 GPT-4V ↑22 63.1%
- 6
Gemini 1.0 Ultra ↑18 62.3%
- 7
Qwen-VL-Plus ↑28 59%
- 8
InternLM-XComposer2-VL-7B ↑15 59%
- 9
InternVL-Chat-V1-2-Plus ↑10 55.7%
- 10
Gemini 1.0 Pro ↑22 54.9%
Statistical
15 models we track have a score here.
- 1
Doubao-1.5-pro 86.5%
- 2
Ovis2-34B 85%
- 3
InternVL2-Pro ↑12 81.1%
- 4
InternVL2-8B-MPO ↑9 79.1%
- 5
InternLM-XComposer2-VL-7B ↑18 64.1%
- 6
InternVL-Chat-V1-2-Plus ↑13 60.8%
- 7
Gemini 1.0 Ultra ↑17 59.5%
- 8
Gemini 1.0 Pro ↑24 56.8%
- 9
Qwen-VL-Plus ↑26 56.1%
- 10 GPT-4V ↑17 55.8%
13 models have held number one in labs’ own numbers since its release in October 2023 14 models have held number one across all independent runs since its release in October 2023 20 models have held number one with claims included since its release in October 2023
Lab claims count from the day the card came out, independent runs from the model’s release. We draw one row each for the benchmark’s own leaderboard and what labs claim in their cards.
Sources: Official MathVista leaderboard and labs’ model cards
OpenAI’s own number for o3 passed 85% in April 2025
That was 18 months after its release. By the same age, 29% of the benchmarks from 2023 had passed 85%. All the scores we have, one dot each, so a model tested at several settings appears more than once.
Sources: Official MathVista leaderboard and labs’ model cards
2 labs put it in a launch post in the last six months
Launch posts from every lab that has listed it, one tick per post. Tall colored ticks include it, and short gray ones leave it out.
| Date | Launch post | Score | As the post writes it |
|---|---|---|---|
| Oct 7, 2026 | 88 | MathVista | |
| Aug 12, 2026 | LFM2.5-VL-3B | 68.5 | MathVista (mini) |
| May 25, 2026 | 82.9 | Mathvista (mini) | |
| May 20, 2026 | 80.6 | MathVista | |
| Apr 28, 2026 | 82.8 | MathVista_MINI | |
| Apr 22, 2026 | 87.4 | MathVista mini | |
| Apr 15, 2026 | 86.4 | Mathvista(mini) | |
| Apr 9, 2026 | 85 | MathVista (mini) | |
| Mar 30, 2026 | 86.1 | Mathvista (mini) | |
| Mar 4, 2026 | 75.2 | MathVista_MINI (default mode) | |
| Mar 2, 2026 | 62.2 | Mathvista(mini) 62.2/58.6 (with CI / without CI) | |
| Mar 2, 2026 | 76.7 | Mathvista(mini) 76.7/73.9 (with CI / without CI) | |
| Mar 2, 2026 | 85.1 | Mathvista(mini) | |
| Mar 2, 2026 | 85.7 | Mathvista(mini) | |
| Feb 24, 2026 | 86.2 | Mathvista(mini) | |
| Feb 24, 2026 | 87.8 | Mathvista(mini) | |
| Feb 24, 2026 | 87.4 | Mathvista(mini) | |
| Feb 16, 2026 | 90.3 | Mathvista(mini) | |
| Feb 14, 2026 | 89.8 | MathVista | |
| Feb 6, 2026 | 84.8 | MathVista | |
| Jan 27, 2026 | Kimi K2.5 | 90.1 | MathVista (mini) |
| Jan 13, 2026 | 84 | MathVista (SeRe) | |
| Dec 18, 2025 | 87.7 | MathVista | |
| Dec 8, 2025 | GLM-4.6V | 85.2 | MathVista |
| Nov 11, 2025 | 82.5 | MathVista | |
| Oct 31, 2025 | 77.9 | MathVista mini | |
| Oct 28, 2025 | 76.9 | MathVista* | |
| Sep 23, 2025 | 85.8 | MathVista_mini | |
| Sep 22, 2025 | 75.9 | MathVista_mini | |
| Aug 25, 2025 | 82.7 | MathVista | |
| Aug 12, 2025 | LFM2-VL-1.6B | 51.1 | MathVista |
| Aug 11, 2025 | GLM-4.5V | 84.6 | MathVista |
| Jul 31, 2025 | 73.5 | MathVista | |
| Jul 24, 2025 | 81.5 | MathVista | |
| Jul 1, 2025 | GLM-4.1V-9B-Thinking | 80.7 | MathVista |
| Jun 21, 2025 | Kimi-VL-A3B-Thinking-2506 | 80.1 | MathVista_MINI (Pass@1) |
| Jun 20, 2025 | 67.1 | Mathvista | |
| May 20, 2025 | 73.1 | MathVista | |
| May 12, 2025 | 85.6 | MathVista | |
| Apr 16, 2025 | o4-mini | 84.3 | MathVista |
| Apr 16, 2025 | o3 | 86.8 | MathVista Visual Math Reasoning |
| Apr 14, 2025 | GPT-4.1 nano | 56.2 | MathVista |
| Apr 14, 2025 | GPT-4.1 mini | 73.1 | MathVista |
| Apr 14, 2025 | GPT-4.1 | 72.2 | MathVista |
| Apr 10, 2025 | Kimi-VL-A3B-Thinking | 71.3 | MathVista (mini) (Pass@1) |
| Apr 5, 2025 | 70.7 | MathVista | |
| Apr 5, 2025 | 73.7 | MathVista | |
| Mar 27, 2025 | 67.9 | MathVistatestmini | |
| Mar 17, 2025 | 68.9 | Mathvista | |
| Mar 16, 2025 | MathVista | ||
| Mar 12, 2025 | 67.6 | MathVista (testmini) (IT) | |
| Feb 26, 2025 | 62.4 | MathVista (testmini) | |
| Jan 28, 2025 | 74.8 | MathVista_MINI | |
| Jan 20, 2025 | Kimi k1.5 | 74.9 | MathVista-Test (Pass@1), long-CoT |
| Dec 17, 2024 | o1 | 71 | MathVista (pass @1) |
| Oct 22, 2024 | Claude 3.5 Sonnet (October 2024) | 70.7 | MathVista (testmini) |
| Sep 12, 2024 | o1-preview | MathVista (testmini) pass@1 | |
| Aug 13, 2024 | Grok-2 | 69 | MathVista (0-shot CoT) |
| Jul 18, 2024 | GPT-4o mini | 56.7 | MathVista |
| Jun 20, 2024 | Claude 3.5 Sonnet | 67.7 | MathVista (testmini) |
| May 14, 2024 | 63.9 | MathViSTA (testmini), 0-shot | |
| May 13, 2024 | GPT-4o | 63.8 | MathVista (%) (testmini) |
| Mar 4, 2024 | Claude 3 Opus | 50.5 | MathVista (testmini) CoT |
| Feb 15, 2024 | 52.1 | MathViSTA (testmini), 0-shot | |
| Dec 6, 2023 | 53 | MathVista (testmini) |
42 system cards report it, from 17 labs
Besides launch posts, we read the system cards, model cards and technical reports labs publish with a release. They report far more, including safety and frontier-risk tests.
| Date | Document | Score | As the document writes it | Section |
|---|---|---|---|---|
| Jun 23, 2026 | | 90.7 | MathVista (Table 9) | Capabilities |
| Jun 9, 2026 | | 82.9 | MathVista (mini) (Figure 5, printed value) | Capabilities |
| Apr 27, 2026 | | 82.8 | MathVista-Mini, reasoning on, 71.9 reasoning off (Table 7) | Capabilities |
| Apr 17, 2026 | | 86.1 | Mathvista (mini) (Table 6) | Capabilities |
| Apr 9, 2026 | | 85 | MathVista (mini) (Table 2, Reasoning mode) | Capabilities |
| Mar 21, 2026 | | 87.7 | MathVista (Table 2) | Capabilities |
| Mar 4, 2026 | | 75.2 | MathVista MINI, default mode, 68.7 force nothink, 74.1 force thinking (Tables 5 and 6) | Capabilities |
| Feb 14, 2026 | | 89.8 | MathVista (testmini) (Table 8) | Capabilities |
| Feb 4, 2026 | | 84.8 | MathVista (Table 3) | Capabilities |
| Jan 30, 2026 | Kimi K2.5 technical report | 90.1 | MathVista (mini) (Table 4) | Capabilities |
| Jan 14, 2026 | | 84 | MathVista, SeRe (Tables 1 and 3). PaCoRe 85.50 | Capabilities |
| Jan 1, 2026 | GLM-4.6V technical report | 85.2 | MathVista, GLM-4.6V thinking (Table 2) | Capabilities |
| Nov 28, 2025 | LFM2-VL-1.6B technical report | 51.7 | MathVista, LFM2-VL-1.6B (Table 8) | Capabilities |
| Nov 26, 2025 | | 85.8 | MathVista mini (Table 2) | Capabilities |
| Nov 6, 2025 | | 75.5 | MathVista-Mini, reasoning on (Table 3). Reasoning off 69.0 | Capabilities |
| Oct 31, 2025 | | 77.9 | MathVista mini (Table 5) | Capabilities |
| Sep 22, 2025 | | 75.9 | MathVista mini, Instruct (Table 9). Thinking 80.0 (Table 10) | Capabilities |
| Aug 25, 2025 | | 82.7 | MathVista (mini) (Table 2, Table 3). With parallel thinking 84.8 | Capabilities |
| Aug 21, 2025 | | 81.5 | MathVista (Table 2) | Capabilities |
| Aug 13, 2025 | GLM-4.5V technical report | 84.6 | MathVista, GLM-4.5V thinking (Table 2) | Capabilities |
| Jul 1, 2025 | GLM-4.1V-9B-Thinking technical report | 80.7 | MathVista, GLM-4.1V-9B-Thinking thinking (Table 2) | Capabilities |
| Jun 23, 2025 | Kimi-VL-A3B-Thinking-2506 technical report | 80.1 | MathVista (mini) (Pass@1) (Table 4) | Capabilities |
| Jun 12, 2025 | | 66.8 | MathVista, Magistral Small bar, printed label (Figure 10) | Capabilities |
| Jun 12, 2025 | | 70.1 | MathVista, printed bar label (Figure 10) | Capabilities |
| May 20, 2025 | | 73.1 | MathVista (Table 4) | Capabilities |
| May 11, 2025 | | 85.6 | MathVista, thinking (Table 6). Non-thinking 83.0 | Capabilities |
| Apr 10, 2025 | Kimi-VL-A3B-Thinking technical report | 71.3 | MathVista (mini) (Pass@1) (Table 4). Figure 13 shows 70.9 at 4k max thinking tokens | Capabilities |
| Apr 5, 2025 | | 70.7 | MathVista, 0-shot, accuracy, instruction tuned (Instruction tuned models table) | Capabilities |
| Apr 5, 2025 | | 73.7 | MathVista, 0-shot, accuracy, instruction tuned (Instruction tuned models table) | Capabilities |
| Mar 26, 2025 | | 67.9 | MathVista testmini (Table 5) | Capabilities |
| Mar 25, 2025 | | 50 | MathVista (testmini), IT, P&S applied (Table 16) | Capabilities |
| Mar 25, 2025 | | 67.6 | MathVista (testmini), IT, P&S applied (Table 16) | Capabilities |
| Mar 25, 2025 | | 62.9 | MathVista (testmini), IT, P&S applied (Table 16) | Capabilities |
| Mar 3, 2025 | | 62.4 | MathVista (testmini) (Table 1) | Capabilities |
| Feb 19, 2025 | | 74.8 | MathVista mini (Table 3) | Capabilities |
| Jan 20, 2025 | Kimi k1.5 technical report | 74.9 | MathVista-Test (Pass@1), long-CoT (Table 2). Short-CoT 70.1 (Table 3) | Capabilities |
| Oct 22, 2024 | Claude 3.5 Sonnet (October 2024) model card | 70.7 | MathVista (testmini), 0-shot CoT (Table 5) | Capabilities |
| Aug 30, 2024 | | 43.9 | MathVista (testmini) (Table 5) | Capabilities |
| Jun 20, 2024 | Claude 3.5 Sonnet model card | 67.7 | MathVista (testmini), 0-shot CoT (Table 2) | Capabilities |
| Mar 4, 2024 | Claude 3 Sonnet model card | 47.9 | MathVista (testmini), 0-shot CoT (Table 3) | Capabilities |
| Mar 4, 2024 | Claude 3 Opus model card | 50.5 | MathVista (testmini), 0-shot CoT (Table 3) | Capabilities |
| Mar 4, 2024 | Claude 3 Haiku model card | 46.4 | MathVista (testmini), 0-shot CoT (Table 3) | Capabilities |
14 lab claims next to the leaderboard
Each score a lab reported for its own model, beside the same model's best score on the leaderboard. Gaps mostly come from different harnesses, effort levels and task subsets. 13 of them match, which usually means one side copied the other. Every benchmark.
| Model | Lab says | Board says | Gap |
|---|---|---|---|
| o1 | 71% MathVista (pass @1) | 73.9% Leaderboard | −2.9 pts |
| | 53% MathVista (testmini) | 53% Leaderboard | Same |
| | 52.1% MathViSTA (testmini), 0-shot | 52.1% Leaderboard | Same |
| Claude 3 Opus | 50.5% MathVista (testmini) CoT | 50.5% Leaderboard | Same |
| GPT-4o May 2024 | 63.8% MathVista (%) (testmini) | 63.8% Leaderboard | Same |
| | 63.9% MathViSTA (testmini), 0-shot | 63.9% Leaderboard | Same |
| Claude 3.5 Sonnet | 67.7% MathVista (testmini) | 67.7% Leaderboard | Same |
| Grok-2 Aug 2024 | 69% MathVista (0-shot CoT) | 69% Leaderboard | Same |
| Kimi k1.5 | 74.9% MathVista-Test (Pass@1), long-CoT | 74.9% Leaderboard | Same |
| | 62.4% MathVista (testmini) | 62.4% Leaderboard | Same |
| | 73.7% MathVista | 73.7% Leaderboard | Same |
| Claude 3 Haiku | 46.4% MathVista (testmini), 0-shot CoT (Table 3) | 46.4% Leaderboard | Same |
| Claude 3 Sonnet | 47.9% MathVista (testmini), 0-shot CoT (Table 3) | 47.9% Leaderboard | Same |
| Phi-3.5-vision-instruct | 43.9% MathVista (testmini) (Table 5) | 43.9% Leaderboard | Same |
More in Vision
- MMMU In use 11.5K college-level multimodal questions across 30 subjects. Multiple-choice and open accuracy.
Qwen 3.6 Plus 86%
- MMMU-Pro In use Harder MMMU: college-level multimodal questions with more options and vision-only inputs to remove text shortcuts. GPT-5.6 Sol 84.6%
- CharXiv In use Chart understanding on real arXiv figures. Reasoning questions require synthesizing information from complex charts. Claude Fable 5 93.5%
- MATH-Vision (MathVision) In use 3,040 competition math problems with visual context. Answer accuracy.
Qwen 3.8 Max 97.7%
- MMStar In use Models answer 1,500 human-curated image questions that cannot be solved without the image, across 6 core capabilities. Scored by accuracy.
Qwen 3.8 Max 85.9%
- ChartQA In use Models answer human-written and generated questions about chart images that need visual and arithmetic reasoning. Scored by relaxed accuracy with a 5 percent numeric tolerance.
Command A Vision 90.9%
Data audit
All 10 checks pass. Every benchmark
Catalog record 2 pass
- Pass Task access All tasks are public.
- Pass Confidence Checked against a primary source.
Scores 3 pass 2 notes
- Pass Independent runs Scores from the official leaderboard (78 entries), launch posts (63 scores) and system cards (42 scores).
- Pass Models scored 100 models we track, plus 35 other entries on its leaderboard.
- Pass Newest score The newest is from October 2026, for MiniCPM-V-4.7-35B-A3B.
- Note Frontier models None of the top 10 has a score, which is fine for a retired test. The strongest tested is Kimi K2.5 (#76).
- Note Headroom Saturated. The best is 90.7%, past 85% since April 2025.
Leaderboard import 3 pass
- Pass Importer We read the board on mathvista.github.io: 45 rows for models we track and 33 other entries at the last read.
- Pass Last read November 18th, 2026.
- Pass Read worked The last read worked.
Consistency 2 pass
- Pass Lab claims All 14 pairs agree within 10 points.
- Pass Known issues No published audit found problems.