Skip to content

feat: add llmman provider - #3880

Open
ericcurtin wants to merge 1 commit into
chatboxai:mainfrom
ericcurtin:feat/llmman-provider
Open

feat: add llmman provider#3880
ericcurtin wants to merge 1 commit into
chatboxai:mainfrom
ericcurtin:feat/llmman-provider

Conversation

@ericcurtin

Copy link
Copy Markdown

Description

Adds llmman, which runs local models distributed as OCI artifacts and serves an OpenAI-compatible API on port 17434 (it also speaks the Ollama and Anthropic APIs, but Chatbox only needs the OpenAI one).

Because the API is OpenAI-compatible, the definition and model class mirror lmstudio.ts rather than ollama.tsLlmman extends OpenAICompatible, same as LMStudio. It's registered alongside the other local providers in the places that treat a provider as usable without an API key: useProviders, settingActions, llm_utils, and the $providerId route list.

Additional Notes

No provider icon is included. ProviderIcon has no branch for llmman, so it falls through to the empty <svg> and renders blank. Only Custom, OpenAIResponses and VolcEngine are in that position today, so this is a visible gap rather than the norm. I didn't want to guess at a 24×24 monochrome mark for someone else's logo — happy to add one if you tell me what you'd accept, or to take a path you provide.

Testing: tsc --noEmit reports 6005 errors both before and after my change, verified against a stashed baseline. That baseline is high because I typechecked without installing dependencies, so most of it is module resolution — it's the delta that's meaningful here, not the absolute number. An earlier iteration showed a delta of 2, which is how I found a missing aiProviderNameHash entry in src/shared/models/index.ts; both are fixed and the delta is now zero.

Not verified: no run of the app against a live server, so the settings UI and a real completion are untested.

AI-assisted (OpenCode), reviewed before submitting.

Contributor Agreement

I've deliberately left the box below unticked: it's a rights grant that the account owner should make personally rather than have tooling tick on their behalf. @ericcurtin will confirm it in a comment below.

[ ] I have read and agree with the above statement.

llmman (https://github.com/llmmanorg/llmman) runs local models distributed
as OCI artifacts and serves an OpenAI-compatible API on port 17434.

Since the API is OpenAI-compatible, the definition and model class mirror
the LM Studio provider rather than the Ollama one. It is registered
alongside the other local providers in the places that treat a provider as
usable without an API key.

No provider icon is included; ProviderIcon has no branch for llmman and
falls through to the empty svg.

Signed-off-by: Eric Curtin <eric.curtin@docker.com>
@coderabbitai

coderabbitai Bot commented Aug 31, 2026

Copy link
Copy Markdown
Contributor

Important

Review skipped

Auto reviews are disabled on this repository. Please check the settings in the CodeRabbit UI or the .coderabbit.yaml file in this repository. To trigger a single review, invoke the @coderabbitai review command.

⚙️ Run configuration

Configuration used: Repository UI

Review profile: CHILL

Plan: Pro Plus

Run ID: 181b536d-48c3-4acc-9e4e-269634d7fb32

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

Use the checkbox below for a quick retry:

  • 🔍 Trigger review

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

Comment @coderabbitai help to get the list of available commands.

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.

1 participant