
Mastering LLM Quantization: Architecture and Code
From Quantization Mathematics to CUDA, Triton, Inference Runtimes, and Production Optimization
Sammendrag
Om denne boken
- Språk:English
- ISBN:9781808821608
- Utgivelsesdato:18. des. 2026
- Forlag:Packt Publishing
- Trykk:Packt Publishing