server: add presets (config) when using multiple models #17859
aldehir
dismissed these changes
on 2025-12-08
ngxson
commented
on 2025-12-08
ngxson
commented
on 2025-12-08
ngxson
commented
on 2025-12-08
ngxson
commented
on 2025-12-08
ngxson
commented
on 2025-12-08
ngxson
commented
on 2025-12-08
llama-server: recursive GGUF loading
51be1fae
server : router config POC (INI-based per-model settings)
972369e8
server: address review feedback from @aldehir and @ngxson
d564ebf9
server: adopt aldehir's line-oriented PEG parser
193bead2
server: fix CLI/env duplication in child processes
a17f501c
add common/preset.cpp
31cb86a2
fix compile
a7c7aca6
cont
e5c3c471
allow custom-path models
7b962071
add falsey check
b8d8ffee
server: fix router model discovery and child process spawning
0734bbe4
Revert "server: fix router model discovery and child process spawning"
a7baeab4
clarify about "no-" prefix
a70419c0
correct render_args() to include binary path
97de3114
also remove arg LLAMA_ARG_MODELS_PRESET for child
f645e887
add co-author for ini parser code
6bda0d47
also set LLAMA_ARG_HOST
035f56ad
add CHILD_ADDR
f2ad7dc9
Remove dead code
b36b3fe1
ngxson
approved these changes
on 2025-12-10
ngxson
dismissed their stale review
45 days ago
ngxson
changed the title Server: router per model config server: add presets (config) when using multiple models 45 days ago
ngxson
merged
f32ca51b
into master 45 days ago
Assignees
No one assigned