Fix generate.py for fbgemm int4 integration #2273
Merged
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
Summary: Updated table: | | overall tokens/sec | TTFT | Peak Memory | Model Size | | ---------| -------------------| ------| --------------| -----------| | baseline - 1 | 131.65 | 0.0220 | 16.24 GB | 15.01 GB | | baseline - 128| 76.38 | 0.0544 | 26.92 GB | 15.01 GB| | int4wo - 1 | 207.69 | 0.0288 | 6.41 GB | 3.99 GB | | int4wo - 128 | 12.85 | 0.4223 | 16.01 GB | 3.99 GB | | fbgemm-int4 - 1 (w/ compile) | 61.12 | 0.0212 | 7.59 GB | 3.00 GB | | fbgemm-int4 - 128 (w/ compile) | 71.23 | 0.0576 | 16.13 GB | 3.99 GB | Verified that compile works: python generate.py --checkpoint_path $CHECKPOINT_PATH/$MODEL_REPO/model.pth --quantization fbgemm-int4-128 --batch_size 1 --compile ========== Average overall tokens/sec: 61.12 Average decode tokens/sec: 61.5512 s Average TTFT: 0.0212 s Average tokens/sec: 61.12 Average Bandwidth: 243.70 GB/s Peak Memory Usage: 7.59 GB Model Size: 3.99 GB python generate.py --checkpoint_path $CHECKPOINT_PATH/$MODEL_REPO/model.pth --quantization fbgemm-int4-128 --batch_size 128 --compile ========== Average overall tokens/sec: 71.23 Average decode tokens/sec: 72.8871 s Average TTFT: 0.0576 s Average tokens/sec: 71.23 Average tokens/sec including batches 9116.81 Average Bandwidth: 284.00 GB/s Peak Memory Usage: 16.13 GB Model Size: 3.99 GB Test Plan: Reviewers: Subscribers: Tasks: Tags:
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/ao/2273
Note: Links to docs will display an error until the docs builds have been completed. ⏳ No Failures, 4 PendingAs of commit 3263c1d with merge base 4d5f657 ( This comment was automatically generated by Dr. CI and updates every 15 minutes. |
facebook-github-bot
added
the
CLA Signed
jerryzh168
added
the
topic: not user facing
andrewor14
approved these changes
May 29, 2025
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.
Summary:
Updated table:
Verified that compile works:
python generate.py --checkpoint_path $CHECKPOINT_PATH/$MODEL_REPO/model.pth --quantization fbgemm-int4-128 --batch_size 1 --compile
==========
Average overall tokens/sec: 61.12
Average decode tokens/sec: 61.5512 s
Average TTFT: 0.0212 s
Average tokens/sec: 61.12
Average Bandwidth: 243.70 GB/s
Peak Memory Usage: 7.59 GB
Model Size: 3.99 GB
python generate.py --checkpoint_path $CHECKPOINT_PATH/$MODEL_REPO/model.pth --quantization fbgemm-int4-128 --batch_size 128 --compile
==========
Average overall tokens/sec: 71.23
Average decode tokens/sec: 72.8871 s
Average TTFT: 0.0576 s
Average tokens/sec: 71.23
Average tokens/sec including batches 9116.81
Average Bandwidth: 284.00 GB/s
Peak Memory Usage: 16.13 GB
Model Size: 3.99 GB
Test Plan:
Reviewers:
Subscribers:
Tasks:
Tags: