Add GGUF model support with GPU acceleration via llama-cpp-python module #3
Conversation
This file contains hidden or 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
This will be used by data augmentation llms like mistral
Tests gpu running with mistral Q5_K_M
msaligane
approved these changes
Oct 6, 2023
gkielian
pushed a commit
that referenced
this pull request
Aug 17, 2024
Adjust snac whisper script names and paths
gkielian
pushed a commit
that referenced
this pull request
Nov 8, 2024
Refactor to allow for compilation
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
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Outdated suggestions cannot be applied.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.
Adding README.md and scripts for quickly setting up GPU-accelerated python wrapper for running GGUF models (llama.cpp).
Also adds a test script to setup and test quickly by grabbing Mistral 7B and running the example prompt.
Soon we can begin augmentation via methods along the lines of:
https://arxiv.org/abs/2309.09530
Adapting Large Language Models via Reading Comprehension
Which is opened up by Mistral's apache-2.0 license.
This should allow us to further increase the efficacy of our datasets, and reduce the number of parameters required for excellent models.