Please release the lora adapter.
#9
by AlexR7 - opened
Please release the lora adapter so that we can load it in vLLM on the fly, current quantized merged model has performance downgrade.
That would be nice
Please release the lora adapter so that we can load it in vLLM on the fly, current quantized merged model has performance downgrade.
I have the lora adapter for our teich AI Qwen3.5 27b distill posted on my huggingface if you would like to try it https://huggingface.co/EclipseMist/Qwen3.5-27b-Opus-4.6-Distill-LoRa