thek0tyara/llama-cpp-turboquant
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions 4

Convert: Make NVFP4 and MXFP4 HF conversions say NVFP4/MXFP4 instead of BF16 (#20730)

Browse source 
* Corrected convert script for NVFP4 naming and updated gguf constants

* Add mostly_MXFP4 to FileType

Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>

* Update convert_hf_to_gguf.py

Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>

* Update convert_hf_to_gguf.py

Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>

* simplify

* set initial value [no ci]

---------

Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>
This commit is contained in:
Michael Wand 2026-03-21 04:35:21 -07:00 • committed by GitHub
parent 29b28a9824
commit eac9c6ea83
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
2 changed files with 12 additions and 2 deletions
Download patch file Download diff file Expand all files Collapse all files

2
gguf-py/gguf/constants.py
View file

@ -3869,6 +3869,8 @@ class LlamaFileType(IntEnum):
# MOSTLY_Q4_0_8_8 = 35 # removed from gguf files, use Q4_0 and runtime repack
MOSTLY_TQ1_0 = 36 # except 1d tensors
MOSTLY_TQ2_0 = 37 # except 1d tensors
MOSTLY_MXFP4_MOE = 38 # except 1d tensors
MOSTLY_NVFP4 = 39 # except 1d tensors
GUESSED = 1024 # not specified in the model file