Move bitpacking.py to prototype and add uintx_utils.py#4152
Merged
Conversation
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
…nd related code [ghstack-poisoned]
… paths - Remove PlainLayout class from dtypes/utils.py - Delete torchao/dtypes/uintx/plain_layout.py - Remove int8 weight and int8 dynamic activation dispatch from AQT dispatch table - Remove AQT embedding dispatch (used PlainAQTTensorImpl) - Change Int8WeightOnlyConfig default to version=2 (removes v1 AQT path) - Change Int8DynamicActivationInt8WeightConfig default to version=2 (removes v1 AQT path) - Remove PlainLayout from public exports - Update tests to use v2 tensor types [ghstack-poisoned]
- Copy bitpacking.py from torchao/dtypes/uintx/ to torchao/prototype/dtypes/uintx/ (original was deleted in the previous commit) - Add uintx_utils.py with UintxTensor and UintxLayout classes needed by autoround and codebook features until they migrate to the new quantization design [ghstack-poisoned]
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/ao/4152
Note: Links to docs will display an error until the docs builds have been completed. ⏳ No Failures, 3 PendingAs of commit 9d8b941 with merge base 1f90b4d ( This comment was automatically generated by Dr. CI and updates every 15 minutes. |
This was referenced Mar 23, 2026
…ntx_utils.py" - Copy bitpacking.py from torchao/dtypes/uintx/ to torchao/prototype/dtypes/uintx/ (original was deleted in the previous commit) - Add uintx_utils.py with UintxTensor and UintxLayout classes needed by autoround and codebook features until they migrate to the new quantization design [ghstack-poisoned]
- Copy bitpacking.py from torchao/dtypes/uintx/ to torchao/prototype/dtypes/uintx/ (original was deleted in the previous commit) - Add uintx_utils.py with UintxTensor and UintxLayout classes needed by autoround and codebook features until they migrate to the new quantization design [ghstack-poisoned]
…ntx_utils.py" - Copy bitpacking.py from torchao/dtypes/uintx/ to torchao/prototype/dtypes/uintx/ (original was deleted in the previous commit) - Add uintx_utils.py with UintxTensor and UintxLayout classes needed by autoround and codebook features until they migrate to the new quantization design [ghstack-poisoned]
- Copy bitpacking.py from torchao/dtypes/uintx/ to torchao/prototype/dtypes/uintx/ (original was deleted in the previous commit) - Add uintx_utils.py with UintxTensor and UintxLayout classes needed by autoround and codebook features until they migrate to the new quantization design [ghstack-poisoned]
…ntx_utils.py" - Copy bitpacking.py from torchao/dtypes/uintx/ to torchao/prototype/dtypes/uintx/ (original was deleted in the previous commit) - Add uintx_utils.py with UintxTensor and UintxLayout classes needed by autoround and codebook features until they migrate to the new quantization design [ghstack-poisoned]
- Copy bitpacking.py from torchao/dtypes/uintx/ to torchao/prototype/dtypes/uintx/ (original was deleted in the previous commit) - Add uintx_utils.py with UintxTensor and UintxLayout classes needed by autoround and codebook features until they migrate to the new quantization design [ghstack-poisoned]
…ntx_utils.py" - Copy bitpacking.py from torchao/dtypes/uintx/ to torchao/prototype/dtypes/uintx/ (original was deleted in the previous commit) - Add uintx_utils.py with UintxTensor and UintxLayout classes needed by autoround and codebook features until they migrate to the new quantization design [ghstack-poisoned]
- Copy bitpacking.py from torchao/dtypes/uintx/ to torchao/prototype/dtypes/uintx/ (original was deleted in the previous commit) - Add uintx_utils.py with UintxTensor and UintxLayout classes needed by autoround and codebook features until they migrate to the new quantization design [ghstack-poisoned]
…ntx_utils.py" - Copy bitpacking.py from torchao/dtypes/uintx/ to torchao/prototype/dtypes/uintx/ (original was deleted in the previous commit) - Add uintx_utils.py with UintxTensor and UintxLayout classes needed by autoround and codebook features until they migrate to the new quantization design [ghstack-poisoned]
- Copy bitpacking.py from torchao/dtypes/uintx/ to torchao/prototype/dtypes/uintx/ (original was deleted in the previous commit) - Add uintx_utils.py with UintxTensor and UintxLayout classes needed by autoround and codebook features until they migrate to the new quantization design [ghstack-poisoned]
andrewor14
approved these changes
Mar 24, 2026
andrewor14
left a comment
Contributor
There was a problem hiding this comment.
Are there any users for UintxTensor currently?
…ntx_utils.py" - Copy bitpacking.py from torchao/dtypes/uintx/ to torchao/prototype/dtypes/uintx/ (original was deleted in the previous commit) - Add uintx_utils.py with UintxTensor and UintxLayout classes needed by autoround and codebook features until they migrate to the new quantization design [ghstack-poisoned]
- Copy bitpacking.py from torchao/dtypes/uintx/ to torchao/prototype/dtypes/uintx/ (original was deleted in the previous commit) - Add uintx_utils.py with UintxTensor and UintxLayout classes needed by autoround and codebook features until they migrate to the new quantization design [ghstack-poisoned]
…ntx_utils.py" - Copy bitpacking.py from torchao/dtypes/uintx/ to torchao/prototype/dtypes/uintx/ (original was deleted in the previous commit) - Add uintx_utils.py with UintxTensor and UintxLayout classes needed by autoround and codebook features until they migrate to the new quantization design [ghstack-poisoned]
- Copy bitpacking.py from torchao/dtypes/uintx/ to torchao/prototype/dtypes/uintx/ (original was deleted in the previous commit) - Add uintx_utils.py with UintxTensor and UintxLayout classes needed by autoround and codebook features until they migrate to the new quantization design [ghstack-poisoned]
…ntx_utils.py" - Copy bitpacking.py from torchao/dtypes/uintx/ to torchao/prototype/dtypes/uintx/ (original was deleted in the previous commit) - Add uintx_utils.py with UintxTensor and UintxLayout classes needed by autoround and codebook features until they migrate to the new quantization design [ghstack-poisoned]
- Copy bitpacking.py from torchao/dtypes/uintx/ to torchao/prototype/dtypes/uintx/ (original was deleted in the previous commit) - Add uintx_utils.py with UintxTensor and UintxLayout classes needed by autoround and codebook features until they migrate to the new quantization design [ghstack-poisoned]
…ntx_utils.py" - Copy bitpacking.py from torchao/dtypes/uintx/ to torchao/prototype/dtypes/uintx/ (original was deleted in the previous commit) - Add uintx_utils.py with UintxTensor and UintxLayout classes needed by autoround and codebook features until they migrate to the new quantization design [ghstack-poisoned]
- Copy bitpacking.py from torchao/dtypes/uintx/ to torchao/prototype/dtypes/uintx/ (original was deleted in the previous commit) - Add uintx_utils.py with UintxTensor and UintxLayout classes needed by autoround and codebook features until they migrate to the new quantization design [ghstack-poisoned]
…ntx_utils.py" - Copy bitpacking.py from torchao/dtypes/uintx/ to torchao/prototype/dtypes/uintx/ (original was deleted in the previous commit) - Add uintx_utils.py with UintxTensor and UintxLayout classes needed by autoround and codebook features until they migrate to the new quantization design [ghstack-poisoned]
- Copy bitpacking.py from torchao/dtypes/uintx/ to torchao/prototype/dtypes/uintx/ (original was deleted in the previous commit) - Add uintx_utils.py with UintxTensor and UintxLayout classes needed by autoround and codebook features until they migrate to the new quantization design [ghstack-poisoned]
Contributor
Author
I don't think so, this one is not very performant I think |
Contributor
Author
|
@claude review |
|
Claude finished @jerryzh168's task in 1m 24s —— View job Review of PR #4152
Summary: This PR moves FindingsBug: Broken test import
from torchao.dtypes.uintx.bitpacking import pack, pack_cpu, unpack, unpack_cpuSince from torchao.prototype.dtypes.uintx.bitpacking import pack, pack_cpu, unpack, unpack_cpuIt's possible this is intentionally handled in another PR in the ghstack — but as it stands on this branch, the test is broken. Worth confirming. The actual diff in this PR looks correct:
Minor observations (non-blocking):
Overall the change is clean — the only concern is the broken test import noted above. |
…ntx_utils.py" - Copy bitpacking.py from torchao/dtypes/uintx/ to torchao/prototype/dtypes/uintx/ (original was deleted in the previous commit) - Add uintx_utils.py with UintxTensor and UintxLayout classes needed by autoround and codebook features until they migrate to the new quantization design [ghstack-poisoned]
- Copy bitpacking.py from torchao/dtypes/uintx/ to torchao/prototype/dtypes/uintx/ (original was deleted in the previous commit) - Add uintx_utils.py with UintxTensor and UintxLayout classes needed by autoround and codebook features until they migrate to the new quantization design [ghstack-poisoned]
Contributor
Author
|
@claude review again |
|
Claude finished @jerryzh168's task in 1m 0s —— View job Re-review of PR #4152
Summary: This PR makes three changes:
FindingsPrevious issue resolved: The broken test import flagged in the last review has been fixed — No stale references: Searched the entire codebase for any remaining references to Import dependencies verified: The other imports in No issues found. The changes are clean and correct. LGTM. |
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
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.
Stack from ghstack (oldest at bottom):
(original was deleted in the previous commit)
autoround and codebook features until they migrate to the new quantization design