NVFP4 GGUF. No quoted tok/s.
Params
9B
Fastest decode
—
Refusal
Heretic
Setups
0
Runs on
Quoted one-stream decode on real boxes.
No quoted pair yet.
Get it running
- 1. DownloadPuts the weights on this computer.
huggingface-cli download Noobito45/Qwen3.8-9B-heretic-uncensored-NVFP4-GGUF - 2. Start itUse this only if the download is GGUF.
llama-server -hf Noobito45/Qwen3.8-9B-heretic-uncensored-NVFP4-GGUF - Open the weightsHugging Face ↗