Add NVFP4 QAT #2666
Open
Add NVFP4 QAT #2666
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:** This commit adds a QAT flow for NVFP4, following the
numerics in `NVFP4Tensor` closely but without the dtyping casting,
swizzling, and the packing/unpacking. Users can call this flow as follows:
```
from torchao.quantization import quantize_
from torchao.quantization.qat import NVFP4FakeQuantizeConfig, QATConfig
qat_config = QATConfig(
activation_config=NVFP4FakeQuantizeConfig(),
weight_config=NVFP4FakeQuantizeConfig(),
step="prepare",
)
quantize_(model, qat_config)
```
**Test Plan:**
```
python test/quantization/test_qat.py -k test_qat_nvfp4
```
[ghstack-poisoned]
This was referenced Aug 1, 2025
Merged
Merged
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/ao/2666
Note: Links to docs will display an error until the docs builds have been completed. ✅ No FailuresAs of commit 87175e9 with merge base 97b090d ( This comment was automatically generated by Dr. CI and updates every 15 minutes. |
andrewor14
added a commit
that referenced
this pull request
Aug 1, 2025
**Summary:** This commit adds a QAT flow for NVFP4, following the
numerics in `NVFP4Tensor` closely but without the dtyping casting,
swizzling, and the packing/unpacking. Users can call this flow as follows:
```
from torchao.quantization import quantize_
from torchao.quantization.qat import NVFP4FakeQuantizeConfig, QATConfig
qat_config = QATConfig(
activation_config=NVFP4FakeQuantizeConfig(),
weight_config=NVFP4FakeQuantizeConfig(),
step="prepare",
)
quantize_(model, qat_config)
```
**Test Plan:**
```
python test/quantization/test_qat.py -k test_qat_nvfp4
```
ghstack-source-id: fe592ca
Pull Request resolved: #2666
meta-cla
bot
added
the
CLA Signed
andrewor14
added
the
topic: new feature
drisspg
reviewed
Aug 4, 2025
| baseline_out = baseline_model(*x) | ||
| sqnr = compute_error(out, baseline_out).item() | ||
| # Use same SQNR threshold as `test_nvfp4_reconstruction` | ||
| # TODO: why is this 0.0 when `use_per_tensor_scale=True`? |
There was a problem hiding this comment.
That is a seems to be a bug it use_per_tensor_scale should be higher, probably supposed to be 10.0
drisspg
reviewed
Aug 4, 2025
| after the initial fp8 (e4m3) block-wise scaling. | ||
| """ | ||
|
|
||
| use_per_tensor_scale: bool = False |
|
Any numeric studies on how / if this improves quant error? Even if its pretty trivial setup |
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.
Stack from ghstack (oldest at bottom):
Summary: This commit adds a QAT flow for NVFP4, following the
numerics in
NVFP4Tensorclosely but without the dtyping casting,swizzling, and the packing/unpacking. Users can call this flow as follows:
Test Plan: