feat: 2.5x faster model generation (and 2.5x lower model costs) #70
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
moinnadeem
changed the title
|
Thanks for the PR @moinnadeem !! Do you mind elaborate on how did you achieve the 2.5 faster and 2.5 lower cost model? And how did you benchmark it? We're working on testing this out before merging. |
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Outdated suggestions cannot be applied.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.
Hello!
🔥 work on the companion app!
I've been hacking around on low-latency inference recently on Replicate. This PR swaps out the default Vicuna model for a model with ~2.5x faster generation (depending on the sequence length).
It's just a base Replicate model, so we can just swap out the model string: https://replicate.com/moinnadeem/fastervicuna_13b
We could also work on getting these things upstreamed to the main Replicate model too.
WIP:
Things to test: