Skip to content

common : dedup preset and cached model entries in /v1/models - #25131

Merged
ngxson merged 1 commit into
ggml-org:masterfrom
angt:common-dedup-preset-and-cached-model-entries-in-v1-models
Jun 29, 2026
Merged

common : dedup preset and cached model entries in /v1/models#25131
ngxson merged 1 commit into
ggml-org:masterfrom
angt:common-dedup-preset-and-cached-model-entries-in-v1-models

Conversation

@angt

@angt angt commented Jun 29, 2026

Copy link
Copy Markdown
Member

Overview

Dedup preset and cached model entries in /v1/models

Additional information

When a model is present in both the local cache and a preset INI section, /v1/models lists it twice.

This is common with a cached model named xxx-UD-xxx.gguf and a matching preset section [org/model:UD-xxx], the server returns two rows for the same model.

See #24695.

Requirements

Signed-off-by: Adrien Gallouët <angt@huggingface.co>
@angt
angt requested a review from a team as a code owner June 29, 2026 11:39
@angt
angt requested a review from ngxson June 29, 2026 11:40
@ngxson
ngxson merged commit 6f4f53f into ggml-org:master Jun 29, 2026
25 checks passed
turbo-tan pushed a commit to turbo-tan/llama.cpp-tq3 that referenced this pull request Jul 1, 2026
…g#25131)

Signed-off-by: Adrien Gallouët <angt@huggingface.co>
@drsect0r

drsect0r commented Jul 1, 2026

Copy link
Copy Markdown

Quick follow-up question. Why did we uppercase the prefix? I have models listed in preset.ini, but currently the models can't be found because the prefix has changed (to something I didn't specify in preset.ini)

adrianhoehne pushed a commit to adrianhoehne/llama.cpp that referenced this pull request Jul 5, 2026
…g#25131)

Signed-off-by: Adrien Gallouët <angt@huggingface.co>
CowboyTim pushed a commit to aardbeiplantje/llama.cpp that referenced this pull request Jul 21, 2026
…g#25131)

Signed-off-by: Adrien Gallouët <angt@huggingface.co>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants