DeepSeek-V2
ModelEconomical mixture-of-experts language model introducing Multi-head Latent Attention and DeepSeekMoE.
- May 6, 2024
- 128K†
- —
- Yes
- —
- 2
Releases
May 17, 2024DeepSeek-V2-0517 API updateModel update
DeepSeek upgraded deepseek-chat with improved instruction following, JSON output and RAG performance.
May 6, 2024DeepSeek-V2 weights releasedOpen weights
DeepSeek published DeepSeek-V2 base and chat weights with 128K context.
† Illustrative seed value, not verified or live market data.

