vrr/koboldcpp
2
0
Fork 0
mirror of https://github.com/LostRuins/koboldcpp.git synced 2026-05-22 19:47:49 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 1

convert : update mtp related help (#23334)

Browse source 
* update mtp related help

* remove outdated experimental text
This commit is contained in:
Sigbjørn Skjæret 2026-05-19 21:16:58 +02:00 • committed by GitHub
parent ac76808e4d
commit b7393a4d19
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
1 changed files with 3 additions and 3 deletions
Download patch file Download diff file Expand all files Collapse all files

6
convert_hf_to_gguf.py
View file

@ -115,15 +115,15 @@ def parse_args() -> argparse.Namespace:
)
parser.add_argument(
"--mmproj", action="store_true",
help="(Experimental) Export multimodal projector (mmproj) for vision models. This will only work on some vision models. A prefix 'mmproj-' will be added to the output file name.",
help="Export multimodal projector (mmproj) for vision models. This will only work on some vision models. An 'mmproj-' prefix will be added to the output file name.",
)
parser.add_argument(
"--mtp", action="store_true",
help="(Experimental) Export only the multi-token prediction (MTP) head as a separate GGUF, suitable for use as a speculative draft. Output file name will get a '-MTP' suffix.",
help="Export only the multi-token prediction (MTP) head as a separate GGUF, suitable for use as a speculative draft. An 'mtp-' prefix will be added to the output file name.",
)
parser.add_argument(
"--no-mtp", action="store_true",
help="(Experimental) Exclude the multi-token prediction (MTP) head from the converted GGUF. Pair with --mtp on a second run to publish trunk and MTP as two files. Note: the split form duplicates embeddings, so the bundled default is more space-efficient overall.",
help="Exclude the multi-token prediction (MTP) head from the converted GGUF. Pair with --mtp on a second run to publish trunk and MTP as two files. Note: the split form duplicates embeddings, but even though the bundled default is more space-efficient overall, this allows differing quantization which may be more performant.",
)
parser.add_argument(
"--mistral-format", action="store_true",