Back to GalleryBack
Model Comparison
Model Comparison
Compare performance, benchmarks, and characteristics
phi-4-multimodal-instruct
deepinfra
Context131072 tokens
Input Price$0.05 / 1M tokens
Output Price$0.1 / 1M tokens
Loading comparison...
Model Comparison
| Metric |  |  |
|---|---|---|
Pricing | ||
| Input Price | $0.05 / 1M tokens | $2 / 1M tokens |
| Output Price | $0.1 / 1M tokens | $8 / 1M tokens |
Capabilities | ||
| Context Window | 131072 tokens | 1047576 tokens |
| Capabilities | tools | |
| Input type | text, image | text, image |
Category Scores | ||
| Overall Average | 37.2 | 52.2 |
| Academia | 21.9 | 54.9 |
| Marketing | 37.4 | 59.0 |
| Science | 37.1 | 66.1 |
| Vision | 55.1 | 74.8 |
| Writing | 34.6 | 56.6 |
| Finance | N/A | 39.0 |
| Maths | N/A | 34.7 |
| Programming | N/A | 32.2 |
Benchmark Tests | ||
| AIME | 9.3 | 43.7 |
| AA Coding Index | _ | 32.2 |
| AAII | 12.4 | 43.4 |
| AA Math Index | _ | 34.7 |
| GPQA | 31.5 | 66.5 |
| HLE | 4.4 | 4.6 |
| LiveCodeBench | 13.1 | 45.7 |
| MATH-500 | 69.3 | 91.3 |
| MMLU | _ | 90.2 |
| MMLU-Pro | 48.5 | 80.6 |
| MMMU | 55.1 | 74.8 |
| SciCode | 11.0 | 38.1 |
phi-4-multimodal-instruct Advantages
- • Lower input token cost ($0.05 / 1M tokens vs $2 / 1M tokens)
gpt-4.1 Advantages
- • Higher benchmark scores (52.2 vs 37.2)
- • Larger context window (1047576 tokens vs 131072 tokens)