llama.cpp
llama: introduce support for model-embedded sampling parameters
#17120
Merged
Merged
llama: introduce support for model-embedded sampling parameters #17120
taronaeo
merged
17 commits into
ggml-org:master
from
taronaeo:feat/auto-sampling-params
common: introduce auto sampling params from metadata
70f568ad
gguf-py: introduce new kv for conversion scripts
7de014e4
gguf-py: fix formatting
c41bb285
gguf-py: fix more formatting issues
caa7a039
gguf-py: introduce support for reading from generation_config.py
44addceb
gguf-py: simplified gen_config loading
0f8d637c
llama: add support for xtc sampler
6cf39000
chore: formatting
c8845ff5
common: introduce support for general.sampler.sequence
33ddb274
gguf-py: revert test_metadata.py
fd3fa3a4
gguf-py: fix linting
fc91c102
llama: rename sampler to sampling
a2701ec2
llama: sampling_config to user_sampling_config for clarity
954278cf
common: missed updating a variable
19f4c104
llama: move metadata keys to libllama
f58b7582
revert: build-xcframework.sh
6e33d2ca
llama: fix typo b/w temp and temperature
a6494f43
ggerganov
approved these changes
on 2025-11-24
CISC
approved these changes
on 2025-11-24
ORippler
commented
on 2025-12-19
Login to write a write a comment.
Login via GitHub