deepseek-ai/DeepSeek-V2
DeepSeek-V2: A Strong, Economical, and Efficient Mixture-of-Experts Language Model
MIT
Issues
- 1
deepseek v2.5 w4a16在vllm-0.6.3.post1运行失败
#99 opened by cxmt-ai-tc - 2
不能在cursor编辑器上用自定义api是吗?
#97 opened by PeyFon - 0
- 1
- 0
- 0
default temperature of this model
#94 opened by ssk705 - 0
NAN issue using FP16 to load the model
#93 opened by zitgit