DeepSeek releases the efficiency-focused DeepSeek-V2
DeepSeek releases a 236B mixture-of-experts model that combines sparse activation with latent attention to cut training and inference costs.
DE1 English report
MODEL
DeepSeek V2
- Release date:
- 2024-05-07
- Parameters:
- 236B (MoE)
- Legacy modality:
- Text
- Openness:
- Open-source
- Organization:
- DeepSeek
Official links | Chat | Code Plan | Agent tools | API |
|---|---|---|---|---|
| Global | — | — | ||
| Mainland China | — | — |
Related English reports
English reports linked to this entity will appear here as their translations are published.
AI-generated