Side-by-side comparison of Llama 4 Scout (Meta · USA) and Phi-4 (Microsoft · USA) for self-hosted deployment of the open-weight model. Llama 4 Scout is rated conditional; Phi-4 is EU-ready. They part ways on licence: Llama 4 Scout is "Llama community", Phi-4 is "MIT".
| Field | ||
|---|---|---|
| Summary | ||
| Verdict | Conditional Smaller Llama 4 variant: 17B active over 16 experts, multimodal. More self-hostable than Maverick. Same Llama community licence caveats. | EU-ready MIT-licensed 14B from Microsoft Research. Heavy use of synthetic training data is disclosed; English-primary (thin multilingual coverage). Strongest small-model option for permissive-licence EU deployments. |
| Last reviewed | 2026-04-15 | 2026-04-15 |
| Open-weight | ||
| Licence | Llama community | MIT |
| Commercial use | With caps | Yes |
| Training data | Undisclosed | Partial |
| Origin | USA | USA |
| Performance & pricing? | ||
| Quality index | 14/100 | 10/100 |
| Speed | 128 tok/s | 29 tok/s |
| Blended price | $0.29/M | $0.22/M |
| Context window | — | — |
| Evidence | ||
| Sources | ||
No overlapping sources between the two entries.