[WIP] Include GPT Fast in torch.compile nightly benchmark workflow #2857
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
sachanub
changed the title
lxning
reviewed
Dec 20, 2023
| gpt_fast: | ||
| 7b_int4: | ||
| benchmark_engine: "ab" | ||
| url: https://torchserve.pytorch.org/mar_files/gpt_fast_7b_int4.mar |
There was a problem hiding this comment.
please clearly specify the model in the name. eg. Llama-2-7b-hf
| backend_profiling: False | ||
| exec_env: "local" | ||
| processors: | ||
| - "cpu" |
|
@namannandan @lxning What is the work remaining for this PR? |
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Outdated suggestions cannot be applied.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.
Description
Please read our CONTRIBUTING.md prior to creating your first pull request.
The objective of this PR is to include the GPT Fast model with weights corresponding to Llama 7B with int4 quantization.
Steps to download Llama 7B weights in the benchmark host:
Ran a temporary workflow to download weights with the HUGGING_FACE_HUB_TOKEN in the commit 1e6088e
Results of successful run: https://github.com/pytorch/serve/actions/runs/7271384883/job/19811851851?pr=2857
Testing:
Ran benchmark workflow in the commit 31936c9
Results of the successful run: https://github.com/pytorch/serve/actions/runs/7272224847/job/19813999840?pr=2857
Benchmark report file: report.md
Updates in benchmark-ab.py script:
Also updated the
benchmark-ab.pyscript to include-lin theabcommands to allow variable response lengths without counting them as errors (https://httpd.apache.org/docs/2.4/programs/ab.html).