Side-by-side comparison of DeepSeek R1 (DeepSeek · China) and Gemma 4 31B Instruct (Google DeepMind · United States) for self-hosted deployment of the open-weight model. DeepSeek R1 is rated conditional; Gemma 4 31B Instruct is conditional. They part ways on licence: DeepSeek R1 is "MIT", Gemma 4 31B Instruct is "Apache 2.0".
| Field | ||
|---|---|---|
| Summary | ||
| Verdict | Conditional Frontier reasoning model at o1-class performance. MIT licence makes weights legally clean. Same Chinese-origin alignment/supply-chain considerations as DeepSeek V3. Distilled Qwen/Llama versions inherit their base licence. | Conditional Based on published licence terms, Gemma 4 31B ships under pure Apache 2.0 — a notable break from the Gemma Terms of Use used in prior generations — with no prohibited-use carve-outs. US origin carries Schrems-II and CLOUD-Act exposure, and at 30.7B dense the model likely crosses EU AI Act GPAI systemic-risk thresholds that enterprise deployers should document. |
| Last reviewed | 2026-04-15 | 2026-04-17 |
| Open-weight | ||
| Licence | MIT | Apache 2.0 |
| Commercial use | Yes | Unrestricted |
| Training data | Undisclosed | Domain-level summary |
| Origin | China | United States |
| Performance & pricing? | ||
| Quality index | 27/100 | 32/100 |
| Speed | — | — |
| Blended price | $2.36/M | — |
| Context window | — | — |
| Evidence | ||
| Sources | ||
No overlapping sources between the two entries.