Microsoft: phi-4-multimodal-instruct
3 benchmarks
Leaderboard position
#710
Overall rank (of 1122 models)
646
Elo rating
3
Benchmarks
Top 63.3%
Percentile
In Microsoft (of 19 models)
Above: inspatio: world #709
Below: Houssem Fatnasssi: kilma-v1-base #711
Benchmark highlights
Worst benchmark
-(Open ASR Leaderboard)
-6.02
Nearest competitor from another company
The model from another company with the most similar benchmark scores.
appeared as nearest in 1 of 3 benchmarks
1/3
33.3%
1/3
33.3%
1/3
33.3%
| Model | Votes | % of benchmarks |
|---|---|---|
| IBM Granite: granite-speech-3.3-2b | 1 | 33.3% |
| Qwen: qwen2.5-coder-7b-instruct | 1 | 33.3% |
| Qwen: Qwen3.5 0.8B | 1 | 33.3% |
Benchmarks
Predicted scores
Estimated from 25 similar models and the global Elo rating. These are projections, not measured results.
| Benchmark | Predicted | Confidence | Based on |
|---|---|---|---|
| Artificial Analysis Intelligence Index v5.0.1 | ~12 | 50% | — |
| MATH-500 | ~82.8 | 50% | — |
| AIME 2025 | ~59.32 | 50% | — |
| AIME 2026 | ~69.51 | 50% | — |
| HMMT Feb 2026 | ~60.99 | 50% | — |
| GPQA Diamond | ~53.53 | 50% | — |
| MMLU-Pro | ~56.16 | 50% | — |
| Humanity's Last Exam (no tools) | ~7.9 | 50% | — |
| IFEval | ~79.45 | 50% | — |
| IFBench | ~49.92 | 50% | — |
| HumanEval | ~67.7 | 50% | — |
| MBPP | ~69.17 | 50% | — |
| Terminal-Bench v2.1 | ~14 | 50% | — |
| MMLU | ~62.35 | 50% | — |
| SWE-bench Verified | ~32.8 | 50% | — |
| SWE-bench Pro | ~12.14 | 50% | — |
| SciCode | ~23.6 | 50% | — |
| AA-LCR | ~38.99 | 50% | — |
| Humanity's Last Exam (with tools) | ~11.9 | 50% | — |
| GSM8K | ~75.05 | 50% | — |
| -(CoderPPL) | ~-2.2 | 50% | — |
| LiveCodeBench v6 | ~51.21 | 50% | — |
| MMMU-Pro | ~49.04 | 50% | — |
| MATH | ~52.09 | 50% | — |
| BBH | ~60.75 | 50% | — |