Phi-3.5 on MMMU: score, rank and what it means
Phi-3.5 scores 68.8 on MMMU, placing it #328 of 600 models in the index. The benchmark average is 67.1 and the leader is GPT-5.2 Max at 86.9.
Score distribution
The chart shows the top 10 models on MMMU, with Phi-3.5 highlighted. Longer bars mean higher scores.
Top 10 leaderboard
| # | Model | Provider | Context | Input / 1M | Avg score |
|---|---|---|---|---|---|
| 1 | GPT-5.2 Max | OpenAI | 1000000 | $21.39 | 89.2 |
| 2 | GPT-5.1 Reasoning | OpenAI | 400000 | $20.27 | 88.3 |
| 3 | GPT-5.2 | OpenAI | 2000000 | $19.71 | 88.0 |
| 4 | Claude Sonnet 4.5 Reasoning | Anthropic | 400000 | $19.48 | 87.7 |
| 5 | GPT-5.1 Max | OpenAI | 2000000 | $20.50 | 87.9 |
| 6 | Grok 4 Pro | xAI | 1000000 | $19.00 | 85.4 |
| 7 | Claude Opus 4.5 Reasoning | Anthropic | 2000000 | $20.91 | 88.2 |
| 8 | Claude Sonnet 4.5 Pro | Anthropic | 1000000 | $18.73 | 86.7 |
| 9 | GPT-5.2 Pro | OpenAI | 2000000 | $21.32 | 89.6 |
| 10 | GPT-5.2 Vision | OpenAI | 1000000 | $19.03 | 87.4 |
What this score means
On MMMU, Phi-3.5 sits at #328 of 600. The leader, GPT-5.2 Max, scores 86.9, so the gap to the top is 18.1 points. The median model in this index scores 70.0, which puts Phi-3.5 below the midpoint.
Microsoft models average 59.7 on this benchmark. Phi-3.5 is 9.1 points above that provider line, so it is a stronger result than the average Microsoft model.
Among open-weight models, the average is 65.1. Phi-3.5 is 3.7 points above that group. This is the more useful comparison if you are shortlisting within the same tier rather than across the whole index.
Read a 68.8 as a relative signal, not an absolute guarantee. Benchmark scores depend on prompt format, evaluation harness and contamination controls. Use this page to narrow the shortlist, then test the model on your own data.
Price context
At $2.81 per million input tokens, Phi-3.5 is below the median input price of $4.00 across the 600 models scored on MMMU. The models immediately around it in the ranking are shown below so you can compare score and price together.
| Model | Provider | Context | Input / 1M | Avg score |
|---|---|---|---|---|
| DBRX Base | Databricks | 32768 | $2.63 | 69.1 |
| DeepSeek-Coder Base | DeepSeek | 32768 | $3.53 | 74.1 |
| OLMo 2 Chat | Allen Institute for AI | 16384 | $2.94 | 70.6 |
| Llama 3 Modal Base | Modal | 16384 | $3.21 | 72.2 |
| Phi-3.5 | Microsoft | 16384 | $2.81 | 70.9 |
| Doubao 1.5 Pro Mini | ByteDance | 32768 | $8.30 | 72.3 |
| Firesearch Pro | Fireworks AI | 32768 | $7.67 | 70.4 |
Other benchmarks for Phi-3.5
| Benchmark | Score | Rank | Percentile | Relative |
|---|---|---|---|---|
| MMLU-Pro | 76.1 | 304 | 49th | |
| GPQA Diamond | 67.0 | 331 | 45th | |
| SWE-bench Verified | 60.9 | 308 | 49th | |
| HumanEval | 76.2 | 327 | 46th | |
| MATH-500 | 75.4 | 316 | 47th | |
| AIME 2025 | 60.6 | 383 | 36th | |
| LiveCodeBench | 59.8 | 371 | 38th | |
| IFEval | 77.1 | 377 | 37th | |
| TruthfulQA | 74.8 | 336 | 44th | |
| HellaSwag | 82.9 | 323 | 46th | |
| Arena Elo | 1356.7 | 350 | 42th |
FAQ
What score does Phi-3.5 get on MMMU?
Phi-3.5 scores 68.8 on MMMU in this index. The benchmark average is 67.1 and the top score is 86.9.
How does Phi-3.5 rank on MMMU?
It ranks #328 of 600 models, in the 45th percentile. The leader is GPT-5.2 Max at 86.9.
Is Phi-3.5 good at MMMU?
It is below the median for this benchmark, so test it carefully before relying on it for this capability. The result should be read alongside its price and context window.
How is the MMMU score calculated?
MMMU is a multimodal benchmark. Scores in this index are modelled estimates on a 0–100 scale (or Elo for Arena), derived from capability tier and benchmark difficulty. They are not official leaderboard results.
What does Phi-3.5 cost?
Phi-3.5 costs $2.81 per million input tokens and $14.99 per million output tokens.