Phi-4 Instruct on AIME 2025: score, rank and what it means
Phi-4 Instruct scores 74.0 on AIME 2025, placing it #182 of 600 models in the index. The benchmark average is 63.6 and the leader is Claude Opus 4.5 Pro at 91.4.
Score distribution
The chart shows the top 10 models on AIME 2025, with Phi-4 Instruct highlighted. Longer bars mean higher scores.
Top 10 leaderboard
| # | Model | Provider | Context | Input / 1M | Avg score |
|---|---|---|---|---|---|
| 1 | Claude Opus 4.5 Pro | Anthropic | 200000 | $20.98 | 89.1 |
| 2 | GPT-5.2 Pro | OpenAI | 2000000 | $21.32 | 89.6 |
| 3 | GPT-5.1 Pro | OpenAI | 200000 | $19.77 | 87.8 |
| 4 | GPT-5.2 | OpenAI | 2000000 | $19.71 | 88.0 |
| 5 | o4 Reasoning | OpenAI | 1000000 | $19.69 | 87.5 |
| 6 | o4 Pro | OpenAI | 1000000 | $19.37 | 87.3 |
| 7 | Claude Opus 4.5 | Anthropic | 400000 | $19.42 | 87.7 |
| 8 | Gemini 3 Pro Max | Google DeepMind | 2000000 | $20.57 | 87.9 |
| 9 | Claude Opus 4.5 Max | Anthropic | 1000000 | $20.90 | 87.8 |
| 10 | Claude Opus 4.5 Reasoning | Anthropic | 2000000 | $20.91 | 88.2 |
What this score means
On AIME 2025, Phi-4 Instruct sits at #182 of 600. The leader, Claude Opus 4.5 Pro, scores 91.4, so the gap to the top is 17.4 points. The median model in this index scores 67.6, which puts Phi-4 Instruct above the midpoint.
Microsoft models average 54.6 on this benchmark. Phi-4 Instruct is 19.4 points above that provider line, so it is a stronger result than the average Microsoft model.
Among multimodal models, the average is 73.9. Phi-4 Instruct is 0.1 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 74.0 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 $3.79 per million input tokens, Phi-4 Instruct is below the median input price of $4.00 across the 600 models scored on AIME 2025. 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 |
|---|---|---|---|---|
| Mistral Large 3 Chat | Mistral AI | 1000000 | $4.90 | 79.1 |
| Nova Pro Reasoning | Amazon | 200000 | $13.48 | 79.4 |
| Phi-4 | Microsoft | 200000 | $3.96 | 75.4 |
| Llama 3.3 Groq | Groq | 32768 | $3.98 | 76.8 |
| Phi-4 Instruct | Microsoft | 128000 | $3.79 | 75.3 |
| Qwen3-Coder Chat | Alibaba Cloud | 128000 | $4.59 | 78.4 |
| Codestral Chat | Mistral AI | 32768 | $3.89 | 75.8 |
Other benchmarks for Phi-4 Instruct
| Benchmark | Score | Rank | Percentile | Relative |
|---|---|---|---|---|
| MMLU-Pro | 76.6 | 289 | 52th | |
| GPQA Diamond | 73.5 | 186 | 69th | |
| SWE-bench Verified | 62.7 | 263 | 56th | |
| HumanEval | 77.8 | 292 | 51th | |
| MATH-500 | 80.0 | 207 | 66th | |
| LiveCodeBench | 69.8 | 195 | 68th | |
| IFEval | 80.9 | 314 | 48th | |
| TruthfulQA | 78.2 | 251 | 58th | |
| HellaSwag | 83.4 | 310 | 48th | |
| MMMU | 71.6 | 253 | 58th | |
| Arena Elo | 1401.7 | 205 | 66th |
FAQ
What score does Phi-4 Instruct get on AIME 2025?
Phi-4 Instruct scores 74.0 on AIME 2025 in this index. The benchmark average is 63.6 and the top score is 91.4.
How does Phi-4 Instruct rank on AIME 2025?
It ranks #182 of 600 models, in the 70th percentile. The leader is Claude Opus 4.5 Pro at 91.4.
Is Phi-4 Instruct good at AIME 2025?
Yes, relative to the index: it is above the median and inside the top half of scores. The result should be read alongside its price and context window.
How is the AIME 2025 score calculated?
AIME 2025 is a mathematics 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-4 Instruct cost?
Phi-4 Instruct costs $3.79 per million input tokens and $17.11 per million output tokens.