-
Notifications
You must be signed in to change notification settings - Fork 50
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
Ianmacleod/add mistral #307
Conversation
@@ -117,6 +117,8 @@ | |||
"falcon-7b-instruct": "tiiuae/falcon-7b-instruct", | |||
"falcon-40b": "tiiuae/falcon-40b", | |||
"falcon-40b-instruct": "tiiuae/falcon-40b-instruct", | |||
"mistral-7b": "mistralai/Mistral-7B-v0.1", |
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.
Do we need to add additional validation/tests? Also, think we need to update the EGP APIs?
Also we may want to update our Model Zoo docs to have some more granularity, to reflect the fact that not all models are supported for all frameworks.
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.
Oh yes, the EGP APIs will also need to change. And I will add the model-zoo comment and add a supported frameworks column to the docs.
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.
as far as testing, we can add synthetic monitors for completions for the new models. I will take out a new pr for this.
model-engine/model_engine_server/domain/use_cases/llm_model_endpoint_use_cases.py
Show resolved
Hide resolved
@@ -117,6 +117,8 @@ | |||
"falcon-7b-instruct": "tiiuae/falcon-7b-instruct", | |||
"falcon-40b": "tiiuae/falcon-40b", | |||
"falcon-40b-instruct": "tiiuae/falcon-40b-instruct", | |||
"mistral-7b": "mistralai/Mistral-7B-v0.1", |
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.
update L494 about context window?
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.
added this in, also added some logic to make the output dir "mistral_files" and added the max_model_len param
adding mistral endpoint support for completions