MMK12
Models use images to answer school mathematics and science questions. The test split contains 2,000 problems with reference answers.
The authors publish separate subject scores for open and proprietary models. Prime Intellect provides an independent evaluation environment.
- Made by
- MM-EUREKA team
- Released
- April 2025
- Kind
- Visual math and diagrams
- Level
- School
- Scored
- Checked automatically
- Answers
- Short answer
- Tasks
- All public
- Measures
- Capability
- Subject
- Mathematics, Physical sciences
- Language
- English
- Interaction
- Single turn
- Items from
- Exams and contests
- Refresh
- One fixed set
- Run by
- Anyone
- In views
- Vision
- Strongest model tested
- o1 #118 on the capability index
- Last model added
- March 2025
- Scores refreshed
- November 21st, 2026
o1 leads with 73.9% among models released by March 2025 o1 leads with 73.9% among models released by March 2025
Best score per entry from the benchmark's paper, read on November 21st, 2026. A paper’s table doesn’t add newer models. Where a model ran with several settings or harnesses, we keep its best. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders. This view shows only models we track, each at its best.
The newest model with a score here came out in March 2025, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:
- GPT-6.2 Astra
- Claude Fable 5.2
-
Gemini 4 Carbon
- GPT-6 Astra
- and 16 more
- 73.9 o1
- 72.2 MM-Eureka-32B
- 70.5
Qwen2.5-VL-72B
- 66.8
Qwen2.5-VL-32B
- 65.2
Gemini 2.0 Flash
- 64.5 MM-Eureka-7B
- 61.6 InternVL2.5-VL-78B
- 61.5
QVQ-72B-Preview
- 60.6 OpenVLThinker-7
- 58.1 Adora
- 58.0 InternVL2.5-VL-38B
- 55.3 Claude 3.7 Sonnet
- 53.6
Qwen2.5-VL-7B
- 49.9 GPT-4o
- 48.3
InternVL2.5-38B-MPO
- 45.6 InternVL2.5-VL-8B
- 39.8 R1-Onevision
- 34.5
InternVL2.5-8B-MPO
- 1 o1 73.9%
- 2 MM-Eureka-32B 72.2%
- 3
Qwen2.5-VL-72B 70.5%
- 4
Qwen2.5-VL-32B 66.8%
- 5
Gemini 2.0 Flash Feb 2025 65.2%
- 6 MM-Eureka-7B 64.5%
- 7 InternVL2.5-VL-78B 61.6%
- 8
QVQ-72B-Preview 61.5%
- 9 OpenVLThinker-7 60.6%
- 10 Adora 58.1%
- 11 InternVL2.5-VL-38B 58%
- 12 Claude 3.7 Sonnet 55.3%
- 13
Qwen2.5-VL-7B 53.6%
- 14 GPT-4o Mar 2025 49.9%
- 15
InternVL2.5-38B-MPO 48.3%
- 16 InternVL2.5-VL-8B 45.6%
- 17 R1-Onevision 39.8%
- 18
InternVL2.5-8B-MPO 34.5%
Source: The MMK12 paper
- 1 o1 73.9%
- 2
Qwen2.5-VL-72B 70.5%
- 3
Qwen2.5-VL-32B 66.8%
- 4
Gemini 2.0 Flash Feb 2025 65.2%
- 5
QVQ-72B-Preview 61.5%
- 6 Claude 3.7 Sonnet 55.3%
- 7
Qwen2.5-VL-7B 53.6%
- 8 GPT-4o Mar 2025 49.9%
- 9
InternVL2.5-38B-MPO 48.3%
- 10
InternVL2.5-8B-MPO 34.5%
Source: The MMK12 paper
| # | Entry | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | o1 | 73.9% | Dec 2024 | Official leaderboard | |
| 2 | MM-Eureka-32B | 72.2% | Official leaderboard | ||
| 3 | | 70.5% | Jan 2025 | Official leaderboard | |
| 4 | | 66.8% | Mar 2025 | Official leaderboard | |
| 5 | | 65.2% | Feb 2025 | Official leaderboard | |
| 6 | MM-Eureka-7B | 64.5% | Official leaderboard | ||
| 7 | InternVL2.5-VL-78B | 61.6% | Official leaderboard | ||
| 8 | | 61.5% | Dec 2024 | Official leaderboard | |
| 9 | OpenVLThinker-7 | 60.6% | Official leaderboard | ||
| 10 | Adora | 58.1% | Official leaderboard | ||
| 11 | InternVL2.5-VL-38B | 58% | Official leaderboard | ||
| 12 | Claude 3.7 Sonnet | 55.3% | Feb 2025 | Official leaderboard | |
| 13 | | 53.6% | Jan 2025 | Official leaderboard | |
| 14 | GPT-4o Mar 2025 | 49.9% | Mar 2025 | Official leaderboard | |
| 15 | | 48.3% | Dec 2024 | Official leaderboard | |
| 16 | InternVL2.5-VL-8B | 45.6% | Official leaderboard | ||
| 17 | R1-Onevision | 39.8% | Official leaderboard | ||
| 18 | | 34.5% | Dec 2024 | Official leaderboard |
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | o1 | 73.9% | Dec 2024 | Official leaderboard | |
| 2 | | 70.5% | Jan 2025 | Official leaderboard | |
| 3 | | 66.8% | Mar 2025 | Official leaderboard | |
| 4 | | 65.2% | Feb 2025 | Official leaderboard | |
| 5 | | 61.5% | Dec 2024 | Official leaderboard | |
| 6 | Claude 3.7 Sonnet | 55.3% | Feb 2025 | Official leaderboard | |
| 7 | | 53.6% | Jan 2025 | Official leaderboard | |
| 8 | GPT-4o Mar 2025 | 49.9% | Mar 2025 | Official leaderboard | |
| 9 | | 48.3% | Dec 2024 | Official leaderboard | |
| 10 | | 34.5% | Dec 2024 | Official leaderboard |
Rankings by part
We read 4 parts from the benchmark’s paper, each ranking models on one part of the test, such as Mathematics, Physics and Chemistry. o1 leads every one of them. Arrows show how far a model moved from its overall rank among the models we track.
Mathematics
10 models we track have a score here.
- 1 o1 81.6%
- 2
Gemini 2.0 Flash Feb 2025 ↑2 76.8%
- 3
Qwen2.5-VL-72B ↓1 75.6%
- 4
Qwen2.5-VL-32B ↓1 71.6%
- 5
QVQ-72B-Preview 61.4%
- 6
Qwen2.5-VL-7B ↑1 58.4%
- 7 Claude 3.7 Sonnet ↓1 57.4%
- 8 GPT-4o Mar 2025 55.8%
- 9
InternVL2.5-38B-MPO 41.4%
- 10
InternVL2.5-8B-MPO 26.6%
Physics
10 models we track have a score here.
- 1 o1 68.8%
- 2
Qwen2.5-VL-72B 64.8%
- 3
Qwen2.5-VL-32B 59.4%
- 4
QVQ-72B-Preview ↑1 57.4%
- 5
Gemini 2.0 Flash Feb 2025 ↓1 53.6%
- 6 Claude 3.7 Sonnet 53.4%
- 7
Qwen2.5-VL-7B 45.4%
- 8
InternVL2.5-38B-MPO ↑1 42.8%
- 9 GPT-4o Mar 2025 ↓1 41.2%
- 10
InternVL2.5-8B-MPO 25%
Chemistry
10 models we track have a score here.
- 1 o1 71.4%
- 2
Qwen2.5-VL-72B 69.6%
- 3
Qwen2.5-VL-32B 69.6%
- 4
Gemini 2.0 Flash Feb 2025 64.6%
- 5
QVQ-72B-Preview 62.6%
- 6
Qwen2.5-VL-7B ↑1 56.4%
- 7
InternVL2.5-38B-MPO ↑2 55.8%
- 8 Claude 3.7 Sonnet ↓2 55.4%
- 9 GPT-4o Mar 2025 ↓1 47%
- 10
InternVL2.5-8B-MPO 42.4%
Biology
10 models we track have a score here.
- 1 o1 74%
- 2
Qwen2.5-VL-72B 72%
- 3
Qwen2.5-VL-32B 66.6%
- 4
Gemini 2.0 Flash Feb 2025 66%
- 5
QVQ-72B-Preview 64.4%
- 6 GPT-4o Mar 2025 ↑2 55.4%
- 7 Claude 3.7 Sonnet ↓1 55%
- 8
Qwen2.5-VL-7B ↓1 54%
- 9
InternVL2.5-38B-MPO 53.2%
- 10
InternVL2.5-8B-MPO 44%
Number one over time
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: The MMK12 paper
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Source: Official MMK12 leaderboard
No lab model card reports it yet
We read the headline results of 540 launch posts from 57 labs, and none of them list this benchmark.
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%
- MathVista In use Math reasoning in visual contexts (charts, geometry, puzzles) from 31 datasets. Answer accuracy.
Seed2.1 Pro 90.7%
- 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%
Data audit
9 of 10 checks pass, 1 warning. Look first at the Newest score check. The newest came 20 months ago, in March 2025, for GPT-4o. Every benchmark
Catalog record 2 pass
- Pass Task access All tasks are public.
- Pass Confidence Checked against a primary source.
Scores 1 warning 3 pass 1 note
- Pass Independent runs Scores from the benchmark’s paper (18 entries).
- Pass Models scored 10 models we track, plus 8 other entries on its leaderboard.
- Warning Newest score The newest came 20 months ago, in March 2025, for GPT-4o.
- Note Frontier models None of the top 10 has a score. The strongest tested is o1 (#118).
- Pass Headroom The best is 73.9%, under 85%, so it still separates models.
Leaderboard import 3 pass
- Pass Importer We read the board on huggingface.co: 10 rows for models we track and 8 other entries at the last read.
- Pass Last read November 21st, 2026.
- Pass Read worked The last read worked.
Consistency 1 pass 1 note
- Note Lab claims No model has both a lab number and an independent run.
- Pass Known issues No published audit found problems.