Compare models
Pick two or three models to compare license, hardware, and live stats side by side.
| Qwen2.5-Coder 32B Instruct | Phi-4 | |
|---|---|---|
| Org | Alibaba | Microsoft |
| Params | 32B | 14B |
| Modality | Text | Text |
| License | PermissiveApache 2.0 | PermissiveMIT |
| Hardware | Prosumer (24-48GB) | Consumer GPU |
| Released | 2024-11 | 2024-12 |
| Take | Widely regarded as the strongest fully open code model at its size, often cited as close to GPT-4-class coding performance. | A data-quality-first small model that punches far above its 14B size on math and reasoning benchmarks, fully MIT licensed. |
| Downloads | — | — |
| Likes | — | — |