NVFP4 for Qwen3.5-27B
Can you please also quantize Qwen/Qwen3.5-27B in NVFP4 as well? Appreciate it!
Can you please also quantize Qwen/Qwen3.5-27B in NVFP4 as well? Appreciate it!
I uploaded an NVFP4 quant earlier today: https://huggingface.co/osoleve/Qwen3.5-27B-NVFP4-MTP
What is MTP in the name? also can you break down the file to a max of 5GB per part? it's easy to download.
also, what GPU did you use to quantize it?
Can you please also quantize Qwen/Qwen3.5-27B in NVFP4 as well? Appreciate it!
I uploaded an NVFP4 quant earlier today: https://huggingface.co/osoleve/Qwen3.5-27B-NVFP4-MTP
What is MTP in the name? also can you break down the file to a max of 5GB per part? it's easy to download.
Also which GPU did you use to quantize it?
Can you please also quantize Qwen/Qwen3.5-27B in NVFP4 as well? Appreciate it!
I uploaded an NVFP4 quant earlier today: https://huggingface.co/osoleve/Qwen3.5-27B-NVFP4-MTP
What is MTP in the name? also can you break down the file to a max of 5GB per part? it's easy to download.
Also which GPU did you use to quantize it?
also, what GPU did you use to quantize it?
Can you please also quantize Qwen/Qwen3.5-27B in NVFP4 as well? Appreciate it!
I uploaded an NVFP4 quant earlier today: https://huggingface.co/osoleve/Qwen3.5-27B-NVFP4-MTP
What is MTP in the name? also can you break down the file to a max of 5GB per part? it's easy to download.
Also which GPU did you use to quantize it?
MTP is multi-token prediction; quantized on the dgx spark
Can you please also quantize qwen/qwen3.5-35b-a3b in NVFP4 as well? Appreciate it!
Hi, the model.safetensors file is too big. I have tried downloading it twice, it gets stuck due to huge file size. Please break it down into ~4GB files.
MTP is multi-token prediction; quantized on the dgx spark
Hi, the model.safetensors file is too big. I have tried downloading it twice, it gets stuck due to huge file size. Please break it down into ~4GB files. Thanks.
MTP is multi-token prediction; quantized on the dgx spark
Hi, the model.safetensors file is too big. I have tried downloading it twice, it gets stuck due to huge file size. Please break it down into ~4GB files. Thanks.
It's in 4 ~5GB files now instead of one 20GB file