kurcontko
/

clef-NVFP4

compressed-tensors

structured-output

8-bit precision

Model card Files Files and versions

clef-NVFP4 / scripts

Commit History

8-bit lookup tables: 20.0 -> 17.5 GB, same accuracy and speed; faster server (prompt-sized CUDA graphs, --gdn-backend)

1abd566
verified

kurcontko commited on about 6 hours ago

Upload clef-NVFP4 (Clef 27B NVFP4, vLLM plugin + SystemOne server)

6ba50d1
verified

kurcontko commited on 4 days ago