The Local LLM Index / Quantization & Formats / #159

woct0rdho/transformers-qwen3-moe-fused

by woct0rdho · Quantization & Formats · updated 3mo ago

Fused Qwen3 MoE layer for faster training, compatible with Transformers, LoRA, bnb 4-bit quant, Unsloth. Also possible to train LoRA over GGUF

39
momentum
258
stars
16
forks
#159
rank
View on GitHub →