New
lithos-ai/lithos-metal v0.1.2 — an LLM inference engine for Apple silicon
A look at lithos-ai/lithos-metal v0.1.2, its Metal inference and serving structure for Apple silicon, and the limits of its projected performance numbers.
Projects
lithos-ai/lithos-metal
lithos-metal: high-performance LLM inference kernels and serving for Apple silicon.
View on GitHub ↗- stars
- ★ 284
- Forks
- 24
- Language
- Python
- License
- Apache-2.0
- Latest release
- v0.1.2
Posts
1No matching stories.