New
johnfink8/sfnx — CUDA inference engine for three PCIe V100s
johnfink8/sfnx is an inference engine that splits model computation across multiple PCIe graphics cards, and its README publishes run results for a specific three-V100 configuration.
johnfink8/sfnx
johnfink8/sfnx is an inference engine that splits model computation across multiple PCIe graphics cards, and its README publishes run results for a specific three-V100 configuration.
View on GitHub ↗- stars
- ★ 4
- Forks
- 0
- Language
- Python
- License
- MIT
Posts
1No matching stories.