Skip to content

chore: changed whisper models base image to ramalama #2677

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

Merged

Conversation

gastoner
Copy link
Contributor

@gastoner gastoner commented Mar 10, 2025

What does this PR do?

Changes base image for whisper models to ramalama

USING latest

Screenshot / video of UI

What issues does this PR fix or reference?

Closes #2631

How to test this PR?

PR checks and try to create a service

@gastoner gastoner requested review from benoitf, jeffmaury and a team as code owners March 10, 2025 05:24
@gastoner gastoner requested review from feloy and deboer-tim March 10, 2025 05:24
Copy link
Collaborator

@benoitf benoitf left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

it should not be the default ramalama image
this image expects an entrypoint, etc

the image with the "ready-to-go" entrypoint is quay.io/ramalama/ramalama-whisper-server

@gastoner gastoner force-pushed the use_ramalama_base_image_for_whisper_models branch from 67152df to 1914b97 Compare March 10, 2025 11:54
@benoitf benoitf dismissed their stale review March 10, 2025 17:11

changes

@gastoner gastoner force-pushed the use_ramalama_base_image_for_whisper_models branch from 1914b97 to 2593cd6 Compare March 14, 2025 05:17
@jeffmaury
Copy link
Collaborator

See containers/ramalama#964

@jeffmaury jeffmaury marked this pull request as draft March 14, 2025 10:31
@gastoner gastoner force-pushed the use_ramalama_base_image_for_whisper_models branch from 2593cd6 to 9bd4c2a Compare March 17, 2025 05:30
@gastoner
Copy link
Contributor Author

@benoitf @jeffmaury should I use latest tag or "currently latest" (now it is 6.4 but I think will be created re-spin - 6.5 since containers/ramalama#977?)

@jeffmaury
Copy link
Collaborator

@benoitf @jeffmaury should I use latest tag or "currently latest" (now it is 6.4 but I think will be created re-spin - 6.5 since containers/ramalama#977?)

Use latest so that we can test

@benoitf
Copy link
Collaborator

benoitf commented Mar 18, 2025

I think 0.6.4 tag was overrided so should be good

@gastoner gastoner force-pushed the use_ramalama_base_image_for_whisper_models branch from 9bd4c2a to e68f62f Compare March 18, 2025 06:29
@gastoner
Copy link
Contributor Author

Using latest now

@gastoner gastoner requested a review from benoitf March 18, 2025 07:46
@gastoner
Copy link
Contributor Author

gastoner commented Mar 20, 2025

When trying to run the whisper recipe I get Something went wrong while starting the inference server: Error: (HTTP code 500) server error - crun: executable file /usr/bin/whisper-server.sh not found in $PATH: No such file or directory: OCI runtime attempted to invoke a command that was not found Looks like the image has been not updated with the fix so far

@benoitf
Copy link
Collaborator

benoitf commented Mar 20, 2025

could you try with quay.io/fbenoit/ramalama-whisper-server:2025-03-20 (just rebuilt from the main branch of RamaLama so you can check if the upcoming 0.6.5 will work for you)

@gastoner
Copy link
Contributor Author

Yes, on my side it works with your 0.6.5

@benoitf
Copy link
Collaborator

benoitf commented Mar 20, 2025

ok so let's wait monday for the next release :-)

@gastoner
Copy link
Contributor Author

@benoitf @jeffmaury looks like 0.7.0 is out https://quay.io/repository/ramalama/ramalama-whisper-server?tab=tags&tag=latest but I have still the same issue (but Florent's quay.io/fbenoit/ramalama-whisper-server:2025-03-20 worked for me)

@gastoner gastoner force-pushed the use_ramalama_base_image_for_whisper_models branch 4 times, most recently from 77e6936 to ea7349c Compare March 26, 2025 11:15
@gastoner gastoner force-pushed the use_ramalama_base_image_for_whisper_models branch from ea7349c to f6fd637 Compare March 27, 2025 10:09
@gastoner gastoner marked this pull request as ready for review March 27, 2025 10:10
Signed-off-by: Evzen Gasta <[email protected]>
@gastoner gastoner requested a review from a team as a code owner March 27, 2025 10:30
Copy link
Collaborator

@jeffmaury jeffmaury left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM. Tested on Win11Pro

Copy link
Contributor

@axel7083 axel7083 left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Tested on windows, LGTM, was able to use the api

image

@odockal odockal requested a review from ScrewTSW March 27, 2025 12:14
@gastoner gastoner merged commit 0d38434 into containers:main Apr 3, 2025
7 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
Projects
None yet
Development

Successfully merging this pull request may close these issues.

Use ramalama images for whisper models
4 participants