MMMU leaderboard
College-level questions that need both an image and text: charts, diagrams, tables and figures. Claude Fable 5.1 leads with 90.64%.
- Models
- 81
- Leader
- Claude Fable 5.1
- Source
- Vals AI
- As of
- Oct 24, 2026
Open weightsProprietaryBest for the price
Full leaderboard
| Rank | Model | Lab | Accuracy | Cost / test | Harness | Price in / out |
|---|---|---|---|---|---|---|
| 1 | Claude Fable 5.1 | Anthropic | 90.64% | $10 / $50 | — | $10 / $50 |
| 2 | Claude Opus 5 | Anthropic | 89.88% | $5 / $25 | — | $5 / $25 |
| 3 | Claude Fable 5 | Anthropic | 89.31% | $10 / $50 | — | $10 / $50 |
| 4 | Gemini 3.8 Flash | 89.08% | $1.5 / $7.5 | — | $0.82 / $4.13 | |
| 5 | Gemini 3.7 Flash | 88.96% | $1.5 / $7.5 | — | $0.75 / $3.75 | |
| 6 | GPT-5.6 Sol | OpenAI | 88.84% | $4 / $20 | — | $4 / $20 |
| 7 | Gemini 3.6 Flash | 88.38% | $1.5 / $7.5 | — | $0.75 / $3.75 | |
| 8 | Gemini 3.5 Flash | 88.27% | $1.5 / $9 | — | $1.50 / $9 | |
| 9 | GPT-5.5 | OpenAI | 88.27% | $5 / $30 | — | $5 / $30 |
| 10 | Gemini 3.1 Pro Preview | 88.21% | $2 / $12 | — | $2 / $12 | |
| 11 | Kimi K3 | Moonshot AI | 88.15% | $3 / $15 | — | $0.24 / $11 |
| 12 | Qwen3.8-Max | Alibaba | 88.03% | $2 / $6 | — | $2 / $6 |
| 13 | Gemini 3 Flash | 87.63% | $0.5 / $3 | — | $0.50 / $3 | |
| 14 | Gemini 3 Pro | 87.51% | $2 / $12 | — | — | |
| 15 | GPT-5.4 | OpenAI | 87.51% | $2.5 / $15 | — | $2.50 / $15 |
| 16 | Muse Spark | Meta | 87.40% | N/A | — | — |
| 17 | GPT-5.2 | OpenAI | 86.67% | $1.75 / $14 | — | $1.75 / $14 |
| 18 | Muse Spark 1.1 | Meta | 86.59% | $1.25 / $4.25 | — | $1.25 / $4.25 |
| 19 | Claude Opus 4.8 | Anthropic | 86.59% | $5 / $25 | — | $5 / $25 |
| 20 | GPT-5.6 Terra | OpenAI | 86.47% | $2 / $12 | — | $2 / $12 |
| 21 | Kimi K2.6 | Moonshot AI | 86.30% | $0.95 / $4 | — | $0.65 / $3.41 |
| 22 | Muse Spark 1.2 | Meta | 86.13% | $1.25 / $4.25 | — | $1.25 / $4.25 |
| 23 | GLM-5.3-Flash | Z.ai | 86.01% | $0.075 / $0.25 | — | $0.15 / $0.50 |
| 24 | Claude Opus 4.7 | Anthropic | 85.55% | $5 / $25 | — | $5 / $25 |
| 25 | GPT-5.6 Luna | OpenAI | 85.03% | $0.2 / $1.2 | — | $0.20 / $1.20 |
| 26 | Kimi K2.5 | Moonshot AI | 84.33% | $0.6 / $3 | — | $0.50 / $2.50 |
| 27 | Qwen3.6-Plus | Alibaba | 84.16% | $0.5 / $3 | — | $0.33 / $1.95 |
| 28 | Qwen3.8-27B | Alibaba | 83.93% | $0.5 / $3 | — | $0.20 / $2.50 |
| 29 | Claude Opus 4.6 | Anthropic | 83.87% | $5 / $25 | — | $5 / $25 |
| 30 | Gemini 3.5 Flash-Lite | 83.58% | $0.3 / $2.5 | — | $0.30 / $2.50 | |
| 31 | Claude Sonnet 4.6 | Anthropic | 83.58% | $3 / $15 | — | $3 / $15 |
| 32 | Grok 4.20 | xAI | 83.47% | $2 / $6 | — | $1.25 / $2.50 |
| 33 | GPT-5.1 | OpenAI | 83.18% | $1.25 / $10 | — | $1.25 / $10 |
| 34 | Grok 4.3 | xAI | 83.06% | $1.25 / $2.5 | — | $1.25 / $2.50 |
| 35 | Claude Sonnet 5 | Anthropic | 83.01% | $2 / $10 | — | $2 / $10 |
| 37 | Gemini 3.1 Flash-Lite Preview | 82.49% | $0.25 / $1.5 | — | $0.25 / $1.50 | |
| 38 | Qwen3.5-Flash | Alibaba | 81.91% | $0.1 / $0.4 | — | $0.065 / $0.26 |
| 39 | GPT-5 | OpenAI | 81.50% | $1.25 / $10 | — | $1.25 / $10 |
| 40 | Gemini 2.5 Pro | 81.34% | $1.25 / $10 | — | $1.25 / $10 | |
| 41 | MiniMax-M3 | MiniMax | 81.16% | $0.6 / $2.4 | — | $0.30 / $1.20 |
| 42 | Claude Opus 4.5 | Anthropic | 81.10% | $5 / $25 | — | $5 / $25 |
| 44 | o3 | OpenAI | 80.42% | $2 / $8 | — | $2 / $8 |
| 45 | MiMo-V2.5 | Xiaomi | 80.00% | $0.14 / $0.28 | — | $0.12 / $0.24 |
| 46 | o4-mini | OpenAI | 79.67% | $1.1 / $4.4 | — | $1.10 / $4.40 |
| 48 | Claude Sonnet 4.5 | Anthropic | 79.31% | $3 / $15 | — | $3 / $15 |
| 49 | GPT-5.4 mini | OpenAI | 79.25% | $0.75 / $4.5 | — | $0.75 / $4.50 |
| 50 | GPT-5 mini | OpenAI | 78.91% | $0.25 / $2 | — | $0.25 / $2 |
| 51 | IInkling | Thinky | 78.50% | $1 / $4.05 | — | $1 / $4.05 |
| 53 | o1 | OpenAI | 77.41% | $15 / $60 | — | $15 / $60 |
| 54 | Grok 4 | xAI | 76.27% | $3 / $15 | — | — |
| 58 | Claude Opus 4.1 | Anthropic | 73.72% | $15 / $75 | — | $15 / $75 |
| 59 | GPT-5.4 nano | OpenAI | 73.58% | $0.2 / $1.25 | — | $0.20 / $1.25 |
| 60 | Claude Opus 4 | Anthropic | 73.31% | $15 / $75 | — | — |
| 62 | Grok 4.1 Fast | xAI | 72.66% | $0.2 / $0.5 | — | — |
| 63 | Gemini 2.5 Flash Preview | 72.54% | $0.1 / $0.4 | — | — | |
| 64 | Claude Sonnet 4 | Anthropic | 72.39% | $3 / $15 | — | — |
| 65 | GPT-4.1 | OpenAI | 72.39% | $2 / $8 | — | $2 / $8 |
| 67 | Llama 4 Maverick | Fireworks AI | 71.69% | $0.22 / $0.88 | — | $0.19 / $0.65 |
| 68 | Claude 3.7 Sonnet | Anthropic | 71.52% | $3 / $15 | — | — |
| 69 | GPT-5 nano | OpenAI | 70.94% | $0.05 / $0.4 | — | $0.05 / $0.40 |
| 70 | GPT-4.1 mini | OpenAI | 70.54% | $0.4 / $1.6 | — | $0.40 / $1.60 |
| 71 | Gemini 2.0 Flash | 69.79% | $0.1 / $0.4 | — | — | |
| 72 | Claude 3.5 Sonnet | Anthropic | 68.80% | $3 / $15 | — | — |
| 73 | Gemini 2.5 Flash Preview 4/17 | 68.00% | $0.3 / $2.5 | — | — | |
| 74 | Mistral Large | Mistral AI | 66.19% | $0.5 / $1.5 | — | $0.50 / $1.50 |
| 75 | Gemini 1.5 Pro | 65.51% | $1.25 / $5 | — | — | |
| 77 | Magistral Medium | Mistral AI | 64.57% | $2 / $5 | — | — |
| 79 | Grok 4.1 Fast Non-Reasoning | xAI | 63.70% | $0.2 / $0.5 | — | — |
| 80 | Grok 4 Fast | xAI | 63.41% | $0.2 / $0.5 | — | — |
| 81 | Mistral Medium | Mistral AI | 62.97% | $0.4 / $2 | — | — |
| 82 | GPT-4o | OpenAI | 62.16% | $2.5 / $10 | — | $2.50 / $10 |
| 83 | Grok 4.5 | xAI | 61.79% | $2 / $6 | — | $2 / $6 |
| 84 | Magistral Small 1.2 | Mistral AI | 60.08% | $0.075 / $0.3 | — | — |
| 85 | Llama 4 Scout | Meta | 58.75% | $0.18 / $0.59 | — | $0.10 / $0.30 |
| 86 | Grok 2 Vision | xAI | 57.25% | $2 / $10 | — | — |
| 87 | Gemini 1.5 Flash | 57.19% | $0.075 / $0.3 | — | — | |
| 88 | GPT-4o mini | OpenAI | 56.56% | $0.15 / $0.6 | — | $0.15 / $0.60 |
| 89 | GPT-4.1 nano | OpenAI | 55.05% | $0.1 / $0.4 | — | $0.10 / $0.40 |
| 91 | Claude Haiku 4.5 | Anthropic | 46.07% | $1 / $5 | — | $1 / $5 |
| 92 | Llama 3.2 Vision | Meta | 38.82% | $0.18 / $0.18 | — | — |
| 93 | Qwen 3.5 Plus | Alibaba | 22.77% | $0.4 / $2.4 | — | $0.30 / $1.80 |
Gemini 4 Argon
Claude Opus 5.5
Muse Wonder
MiMo-V2.6-Pro
Qwen3.8-Max
Kimi K3
GLM-5.3
GPT-5.5
ERNIE 5.1
DeepSeek-V4.1-Flash-20260910
Grok 4.20 Beta
amazon-nova-experimental-chat
Dola Seed 2.0 Pro
NVIDIA-Nemotron-3-Ultra-550B-A55B
Gemma 4 31B
Step 5 Preview
