DeepSeek-V4-Pro-NVFP4 is an NVIDIA-optimized version of the DeepSeek-V4-Pro model using NVFP4 precision. It is hosted on Hugging Face and designed for high-performance inference and commercial AI applications.
In the Quantised & converted weights space, DeepSeek V4 Pro takes a focused approach. It focuses on deploying a high-performance DeepSeek model with NVIDIA-specific optimizations. It is built as an open-source project for developers. The project is open source (Apache-2.0). DeepSeek V4 Pro is available on the web and API.
NVIDIA builds and maintains DeepSeek V4 Pro, and it first shipped in 2024. Development happens publicly on GitHub with 3.3k stars and 355 commits in the last 90 days. It operates in a well-populated space: PulseGate tracks 9 similar projects.
Summary written by a language model from the project’s public pages.
What PulseGate has recorded for this listing
Closest matches by what these projects do