Modelpublic

baa-ai/Qwen3.8-27B-RAM-13GB-GGUF

sourceHugging Faceapache-2.0updated 11h agoView on Hugging Face

3likes343downloads

4 commits on main

51f7a9f11h ago

v2: RAM per-tensor mixed precision (Q2_K-Q8_0 + imatrix) at the same 12.58 GB; WikiText-2 PPL 6.02 vs 6.34 (IQ3_M v1); certified; card updated with 16 GB context-window measurements

tomkay

2dfe1891mo ago

Upload Qwen3.8-27B-RAM-13GB.gguf with huggingface_hub

tomkay

1497a171mo ago

Upload README.md with huggingface_hub

tomkay

e5845ef1mo ago

initial commit

tomkay