Compare models
Phi-4 vs Llama 3.1 8B Instruct vs DeepSeek V3
Every figure below carries the source that reported it and the moment it was observed. Values that failed validation are withheld rather than shown, and a dash means no source we track carries that field for that model.
| Field | Phi-4 | Llama 3.1 8B Instruct | DeepSeek V3 |
|---|---|---|---|
| Downloads, 30 days | 591,256Synced from sourceCurrent3h ago | 6,107,664 (highest of the compared values)Synced from sourceCurrent3h ago | 1,163,331Synced from sourceCurrent3h ago |
| Likes | 2,312Synced from sourceCurrent3h ago | 7,988 (highest of the compared values)Synced from sourceCurrent3h ago | 4,308Synced from sourceCurrent3h ago |
| Context window | 16,384Synced from sourceCurrent3h ago | 131,072Synced from sourceCurrent3h ago | 163,840 (highest of the compared values)Synced from sourceCurrent3h ago |
| Input, $/M tokens | 0.07Synced from sourceCurrent3h ago | 0.05 (lowest of the compared values)Synced from sourceCurrent3h ago | 0.26Synced from sourceCurrent3h ago |
| Modality | textSeeded, unreviewedwritten 1mo ago | textSeeded, unreviewedwritten 1mo ago | textSeeded, unreviewedwritten 1mo ago |
| Licence | mitSeeded, unreviewedwritten 1mo ago | llama-3.1Seeded, unreviewedwritten 1mo ago | deepseekSeeded, unreviewedwritten 1mo ago |
| Gated | noSynced from sourceCurrent3h ago | manualSynced from sourceCurrent3h ago | noSynced from sourceCurrent3h ago |
| Last updated | 2026-07-14Synced from sourceCurrent3h ago | 2024-09-25Synced from sourceCurrent3h ago | 2025-03-27Synced from sourceCurrent3h ago |
| Library | transformersSynced from sourceCurrent3h ago | transformersSynced from sourceCurrent3h ago | transformersSynced from sourceCurrent3h ago |
| Task | text-generationSynced from sourceCurrent3h ago | text-generationSynced from sourceCurrent3h ago | text-generationSynced from sourceCurrent3h ago |
| Hugging Face | microsoft/phi-4Synced from sourceCurrent3h ago | meta-llama/Llama-3.1-8B-InstructSynced from sourceCurrent3h ago | deepseek-ai/DeepSeek-V3Synced from sourceCurrent3h ago |
| OpenRouter id | microsoft/phi-4Synced from sourceCurrent3h ago | meta-llama/llama-3.1-8b-instructSynced from sourceCurrent3h ago | deepseek/deepseek-chatSynced from sourceCurrent3h ago |
| Output, $/M tokens | 0.14Synced from sourceCurrent3h ago | 0.08Synced from sourceCurrent3h ago | 1.03Synced from sourceCurrent3h ago |
| Vendor | MicrosoftSeeded, unreviewedwritten 1mo ago | MetaSeeded, unreviewedwritten 1mo ago | DeepSeekSeeded, unreviewedwritten 1mo ago |
Sources: Hugging Face Hub, OpenRouter model catalogue. Most recent observation 3h ago. Where a row is marked, ▲ is the highest of the values shown and ▼ the lowest — arithmetic on the figures above, not a ranking or a recommendation. Rows where neither extreme is meaningful are left unmarked rather than given a direction they do not have.
Change the comparison
Remove one, or open the directory to pick a different set. Up to 3 at a time.