feat: add candle llama model #19
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Cifko
force-pushed
the
llama-model
branch
4 times, most recently
from
2e0cbaf
to
15285b9
Compare
jorgeantonio21
approved these changes
Apr 4, 2024
| cache: Cache, | ||
| } | ||
|
|
||
| pub struct Input { |
There was a problem hiding this comment.
This needs to conform with Request::ModelInput, as of now. But since this might change in the future, we can merge it for now.
| model_id: Option<String>, | ||
| revision: Option<String>, | ||
| which: Which, | ||
| use_flash_attn: bool, |
There was a problem hiding this comment.
The use_flash_attn is not relevant for fetching. It is a parameter to be used when loading/running the model, to speed up inference runtime by reducing the model weights bandwidth load across different memory components of the GPU.
| }; | ||
| let api = Api::new()?; | ||
| let model_id = fetch.model_id.clone().unwrap_or_else(|| match fetch.which { | ||
| Which::V1 => "Narsil/amall-7b".to_string(), |
There was a problem hiding this comment.
I think we should add an enum ModelId that maps each version to the right string id.
| } | ||
|
|
||
| fn model_id(&self) -> crate::models::ModelId { | ||
| "TinyLlama/TinyLlama-1.1B-Chat-v1.0".to_string() |
There was a problem hiding this comment.
This depends on the model size and should be not hardcoded.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Pr #10 needs to go in first