-
Notifications
You must be signed in to change notification settings - Fork 47
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
fix: adding chatformat to use for inference servers #868
fix: adding chatformat to use for inference servers #868
Conversation
As a general comment I think this is too specific for chat models. We already have the plan to add a type attribute to models (chat, vision,...) so I think it would be better to rename the field to something more general purpose (format ?). And the env variable should be renamed accordingly |
Was thinking intensively about this during the night: found a more generic way to handle that. |
freeform is more extensible 👍 about keys is it using camelCase ? chatFormat -> CHAT_FORMAT ? |
I like this solution but I am not sure about the necessity of the |
"properties" : {
"chatFormat": "openchat"
} ? |
The MODEL_ prefix just to standardize on what is provider by AI Lab as we already provider MODEL_PATH and as it's related to the model |
|
We are using the
podman-desktop-extension-ai-lab/packages/shared/src/models/InferenceServerConfig.ts Line 38 in 039b7ca
|
labels are intended for display I think this is more internal here so I would be more for attributes or properties |
The prefix would not work with containers/podman-desktop-extension-ai-lab-playground-images#11 for now, might want to not having it for now ? |
|
I think that I expect camelCase in a json file also about easier I'm not sure it would be much harder anyway 'chatFormat'.replace(/[A-Z]/g, m => `_${m}`).toUpperCase() => CHAT_FORMAT |
Signed-off-by: axel7083 <42176370+axel7083@users.noreply.github.com>
Signed-off-by: axel7083 <42176370+axel7083@users.noreply.github.com>
Signed-off-by: axel7083 <42176370+axel7083@users.noreply.github.com>
Signed-off-by: axel7083 <42176370+axel7083@users.noreply.github.com>
e40569b
to
bbeeda3
Compare
No problem then |
Signed-off-by: axel7083 <42176370+axel7083@users.noreply.github.com>
I would favor (and not delay) the prefix is the ai-recipes team is ok with this change, starting the discussion on their channel |
Signed-off-by: axel7083 <42176370+axel7083@users.noreply.github.com>
Signed-off-by: axel7083 <42176370+axel7083@users.noreply.github.com>
✅ done |
so here we need to update https://github.com/containers/podman-desktop-extension-ai-lab-playground-images/pull/11/files to use MODEL_ env var |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM. Recipe repo has been updated so we should be good
before merging,
|
Signed-off-by: axel7083 <42176370+axel7083@users.noreply.github.com>
Signed-off-by: axel7083 <42176370+axel7083@users.noreply.github.com>
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM
What does this PR do?
This PR is adding the optional
chatformat
to the model object.Need containers/podman-desktop-extension-ai-lab-playground-images#11 to be merged, and a new version published of the image to be working.
Fixing the playgrounds for the following models
Screenshot / video of UI
What issues does this PR fix or reference?
Fixes #825
How to test this PR?