KYMATATHE LIVING INDEXES
The Living IndexesThe Local LLM Index
The Local LLM Index / Quantization & Formats / #133

sudoingX/qwen38-mtp

by sudoingX · Quantization & Formats · updated 1mo ago

One llama.cpp flag unlocks +33-39% decode speed for Qwen3.8-27B on consumer GPUs. The MTP head already ships inside your GGUF. Recipe, paired benchmarks, probe tool.

57
momentum
367
stars
99
forks
#133
rank
View on GitHub →