Name and Version
ggml_cuda_init: found 1 CUDA devices:
Device 0: NVIDIA GeForce RTX 3090, compute capability 8.6, VMM: yes
version: 8247 (ae87863)
built with GNU 13.3.0 for Linux x86_64
Operating systems
Linux
Which llama.cpp modules do you know to be affected?
llama-server
Command line
llama-server \
--models-preset ./presets.ini \
--models-max 1
Problem description & steps to reproduce
an annoying usability snafu in the webui in that I always have to choose the loaded model because the select doesn't seem to default to one of the models already loaded.
From #18655 (comment)
When using the llama-server router mode, having the Web UI default to at least one of the models (not marked as no-webui) would be welcomed. It appears to know which ones are loaded, but it must be manually selected from the dropdown regardless.
Left: The Web UI initial screen, no model selected despite Qwen 3.5 being loaded.
Right: The dropdown knows that Qwen 3.5 is loaded.
First Bad Commit
No response
Relevant log output
No response
Name and Version
ggml_cuda_init: found 1 CUDA devices:
Device 0: NVIDIA GeForce RTX 3090, compute capability 8.6, VMM: yes
version: 8247 (ae87863)
built with GNU 13.3.0 for Linux x86_64
Operating systems
Linux
Which llama.cpp modules do you know to be affected?
llama-server
Command line
Problem description & steps to reproduce
From #18655 (comment)
When using the llama-server router mode, having the Web UI default to at least one of the models (not marked as no-webui) would be welcomed. It appears to know which ones are loaded, but it must be manually selected from the dropdown regardless.
Left: The Web UI initial screen, no model selected despite Qwen 3.5 being loaded.
Right: The dropdown knows that Qwen 3.5 is loaded.
First Bad Commit
No response
Relevant log output
No response