git
ask
hub
Privacy
Terms
Sign in with GitHub
deepseek-ai/deepseek-v3
↗
Stars ·
103,911
Language ·
Python
License ·
MIT
Simplify and Visualize This Repo
Scan the safety of this repo
Self Host this repo
Check out more work of Developer: deepseek-ai
Ask anything about this repo to start.
Walk me through downloading DeepSeek-V3 weights from Hugging Face and running inference on a multi-GPU server.
How does DeepSeek-V3's Mixture-of-Experts design work, and what does the difference between 671B total and 37B active parameters mean for my hardware requirements?
I want to fine-tune DeepSeek-V3 on a custom dataset, what are the minimum hardware requirements and where do I start?
What quantization options are available for running DeepSeek-V3 on consumer hardware, and which gives the best quality-to-speed trade-off?
Compare DeepSeek-V3's strengths on coding tasks vs GPT-4 so I can decide whether to use it in my project.
Full explanation on explaingit →
📎
Send
By chatting or signing in you agree to the
Terms
and chat-message logging (revocable in
History
).