Deprecate old QAT APIs #2641

andrewor14 · 2025-07-30T22:54:43Z

Stack from ghstack (oldest at bottom):

Summary: Deprecates QAT APIs that should no longer be used.
Print helpful deprecation warning to help users migrate.

Test Plan:

python test/quantization/test_qat.py -k test_qat_api_deprecation

Also manual testing:

>>> from torchao.quantization.qat import IntXQuantizationAwareTrainingConfig
>>> IntXQuantizationAwareTrainingConfig()
'IntXQuantizationAwareTrainingConfig' is deprecated and will be removed in a future release. Please use the following API instead:

    base_config = Int8DynamicActivationInt4WeightConfig(group_size=32)
    quantize_(model, QATConfig(base_config, step="prepare"))
    # train (not shown)
    quantize_(model, QATConfig(base_config, step="convert"))

Alternatively, if you prefer to pass in fake quantization configs:

    activation_config = IntxFakeQuantizeConfig(torch.int8, "per_token", is_symmetric=False)
    weight_config = IntxFakeQuantizeConfig(torch.int4, group_size=32)
    qat_config = QATConfig(
        activation_config=activation_config,
        weight_config=weight_config,
        step="prepare",
    )
    quantize_(model, qat_config)

Please see https://github.com/pytorch/ao/issues/2630 for more details.

IntXQuantizationAwareTrainingConfig(activation_config=None, weight_config=None)

**Summary:** The existing `FakeQuantizeConfig` performs only intx quantization, but we plan to extend QAT to other dtypes such as fp8 and nvfp4 in the near future. This is the necessary refactor before that. Specifically: ``` # New abstract class FakeQuantizeConfigBase # Rename FakeQuantizeConfig -> IntxFakeQuantizeConfig ``` In the future, we will have other types of `FakeQuantizeConfigBase` for float dtypes that users can pass in instead of the existing Intx one. **BC-breaking notes:** For BC, we keep around the old names to reference the new ones. However, this commit is still BC-breaking in the sense that a few APIs now accept the abstract `FakeQuantizeConfigBase` instead. For the most part, this abstract class will be hidden from the user. Before: ``` activation_config = FakeQuantizeConfig(torch.int8, "per_token", is_symmetric=False) weight_config = FakeQuantizeConfig(torch.int4, group_size=32) ``` After: ``` activation_config = IntxFakeQuantizeConfig(torch.int8, "per_token", is_symmetric=False) weight_config = IntxFakeQuantizeConfig(torch.int4, group_size=32) ``` **Test Plan:** python test/quantization/test_qat.py [ghstack-poisoned]

**Summary:** This commit adds a new multi-step QAT API with the main goal of simplifying the existing UX. The new API uses the same `QATConfig` for both the prepare and convert steps, and automatically infers the fake quantization configs based on a PTQ base config provided by the user: ``` from torchao.quantization import ( quantize_, Int8DynamicActivationInt4WeightConfig ) from torchao.quantization.qat import QATConfig \# prepare base_config = Int8DynamicActivationInt4WeightConfig(group_size=32) qat_config = QATConfig(base_config, step="prepare") quantize_(m, qat_config) \# train (not shown) \# convert quantize_(m, QATConfig(base_config, step="convert")) ``` The main improvements include: - A single config for both prepare and convert steps - A single quantize_ for convert (instead of 2) - No chance for incompatible prepare vs convert configs - Much less boilerplate code for most common use case - Simpler config names For less common use cases such as experimentation, users can still specify arbitrary fake quantization configs for activations and/or weights as before. This is still important since there may not always be a corresponding PTQ base config. For example: ``` from torchao.quantization import quantize_ from torchao.quantization.qat import IntxFakeQuantizeConfig, QATConfig activation_config = IntxFakeQuantizeConfig(torch.int8, "per_token", is_symmetric=False) weight_config = IntxFakeQuantizeConfig(torch.int4, group_size=32) qat_config = QATConfig( activation_config=activation_config, weight_config=weight_config, step="prepare", ) quantize_(model, qat_config) \# train and convert same as above (not shown) ``` **BC-breaking notes:** This change by itself is technically not BC-breaking since we keep around the old path, but will become so when we deprecate and remove the old path in the future. Before: ``` \# prepare activation_config = IntxFakeQuantizeConfig(torch.int8, "per_token", is_symmetric=False) weight_config = IntxFakeQuantizeConfig(torch.int4, group_size=32) qat_config = IntXQuantizationAwareTrainingConfig(activation_config, weight_config), quantize_(model, qat_config) \# train (not shown) \# convert quantize_(model, FromIntXQuantizationAwareTrainingConfig()) quantize_(model, Int8DynamicActivationInt4WeightConfig(group_size=32)) ``` After: (see above) **Test Plan:** ``` python test/quantization/test_qat.py ``` [ghstack-poisoned]

**Summary:** This commit adds a new multi-step QAT API with the main goal of simplifying the existing UX. The new API uses the same `QATConfig` for both the prepare and convert steps, and automatically infers the fake quantization configs based on a PTQ base config provided by the user: ``` from torchao.quantization import ( quantize_, Int8DynamicActivationInt4WeightConfig ) from torchao.quantization.qat import QATConfig # prepare base_config = Int8DynamicActivationInt4WeightConfig(group_size=32) quantize_(m, QATConfig(base_config, step="prepare")) # train (not shown) # convert quantize_(m, QATConfig(base_config, step="convert")) ``` The main improvements include: - A single config for both prepare and convert steps - A single quantize_ for convert (instead of 2) - No chance for incompatible prepare vs convert configs - Much less boilerplate code for most common use case - Simpler config names For less common use cases such as experimentation, users can still specify arbitrary fake quantization configs for activations and/or weights as before. This is still important since there may not always be a corresponding PTQ base config. For example: ``` from torchao.quantization import quantize_ from torchao.quantization.qat import IntxFakeQuantizeConfig, QATConfig # prepare activation_config = IntxFakeQuantizeConfig(torch.int8, "per_token", is_symmetric=False) weight_config = IntxFakeQuantizeConfig(torch.int4, group_size=32) qat_config = QATConfig( activation_config=activation_config, weight_config=weight_config, step="prepare", ) quantize_(model, qat_config) # train and convert same as above (not shown) ``` **BC-breaking notes:** This change by itself is technically not BC-breaking since we keep around the old path, but will become so when we deprecate and remove the old path in the future. Before: ``` # prepare activation_config = IntxFakeQuantizeConfig(torch.int8, "per_token", is_symmetric=False) weight_config = IntxFakeQuantizeConfig(torch.int4, group_size=32) qat_config = IntXQuantizationAwareTrainingConfig(activation_config, weight_config), quantize_(model, qat_config) # train (not shown) # convert quantize_(model, FromIntXQuantizationAwareTrainingConfig()) quantize_(model, Int8DynamicActivationInt4WeightConfig(group_size=32)) ``` After: (see above) **Test Plan:** ``` python test/quantization/test_qat.py ``` [ghstack-poisoned]

**Summary:** Deprecates QAT APIs that should no longer be used. Print helpful deprecation warning to help users migrate. **Test Plan:** ``` python test/quantization/test_qat.py -k test_qat_api_deprecation ``` [ghstack-poisoned]

pytorch-bot · 2025-07-30T22:54:47Z

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/ao/2641

📄 Preview Python docs built from this PR

Note: Links to docs will display an error until the docs builds have been completed.

✅ You can merge normally! (2 Unrelated Failures)

As of commit 3f06429 with merge base 5f3ab63 ():

BROKEN TRUNK - The following jobs failed but were present on the merge base:

👉 Rebase onto the `viable/strict` branch to avoid these failures

Run Regression Tests / test-nightly (CPU Nightly, linux.4xlarge, --pre torch --index-url https://download.pytorch.org/wh... / linux-job (gh) (trunk failure)
test/test_low_bit_optim.py::TestFSDP2::test_uneven_shard
Run Regression Tests / test-nightly (CUDA Nightly, linux.g5.12xlarge.nvidia.gpu, --pre torch --index-url https://downloa... / linux-job (gh) (trunk failure)
test/test_low_bit_optim.py::TestFSDP2::test_uneven_shard

This comment was automatically generated by Dr. CI and updates every 15 minutes.

**Summary:** Deprecates QAT APIs that should no longer be used. Print helpful deprecation warning to help users migrate. **Test Plan:** ``` python test/quantization/test_qat.py -k test_qat_api_deprecation ``` ghstack-source-id: f6988d7 Pull Request resolved: #2641

**Summary:** Deprecates QAT APIs that should no longer be used. Print helpful deprecation warning to help users migrate. **Test Plan:** ``` python test/quantization/test_qat.py -k test_qat_api_deprecation ``` Also manual testing: ``` 'IntXQuantizationAwareTrainingConfig' is deprecated and will be removed in a future release. Please use the following API instead: base_config = Int8DynamicActivationInt4WeightConfig(group_size=32) quantize_(model, QATConfig(base_config, step="prepare")) # train (not shown) quantize_(model, QATConfig(base_config, step="convert")) Alternatively, if you prefer to pass in fake quantization configs: activation_config = IntxFakeQuantizeConfig(torch.int8, "per_token", is_symmetric=False) weight_config = IntxFakeQuantizeConfig(torch.int4, group_size=32) qat_config = QATConfig( activation_config=activation_config, weight_config=weight_config, step="prepare", ) quantize_(model, qat_config) Please see #2630 for more details. IntXQuantizationAwareTrainingConfig(activation_config=None, weight_config=None) ``` ghstack-source-id: f6988d7 Pull Request resolved: #2641

**Summary:** Deprecates QAT APIs that should no longer be used. Print helpful deprecation warning to help users migrate. **Test Plan:** ``` python test/quantization/test_qat.py -k test_qat_api_deprecation ``` [ghstack-poisoned]

**Summary:** Deprecates QAT APIs that should no longer be used. Print helpful deprecation warning to help users migrate. **Test Plan:** ``` python test/quantization/test_qat.py -k test_qat_api_deprecation ``` ghstack-source-id: f6988d7 Pull Request resolved: #2641

**Summary:** Deprecates QAT APIs that should no longer be used. Print helpful deprecation warning to help users migrate. **Test Plan:** ``` python test/quantization/test_qat.py -k test_qat_api_deprecation ``` Also manual testing: ``` 'IntXQuantizationAwareTrainingConfig' is deprecated and will be removed in a future release. Please use the following API instead: base_config = Int8DynamicActivationInt4WeightConfig(group_size=32) quantize_(model, QATConfig(base_config, step="prepare")) # train (not shown) quantize_(model, QATConfig(base_config, step="convert")) Alternatively, if you prefer to pass in fake quantization configs: activation_config = IntxFakeQuantizeConfig(torch.int8, "per_token", is_symmetric=False) weight_config = IntxFakeQuantizeConfig(torch.int4, group_size=32) qat_config = QATConfig( activation_config=activation_config, weight_config=weight_config, step="prepare", ) quantize_(model, qat_config) Please see #2630 for more details. IntXQuantizationAwareTrainingConfig(activation_config=None, weight_config=None) ``` [ghstack-poisoned]

**Summary:** Deprecates QAT APIs that should no longer be used. Print helpful deprecation warning to help users migrate. **Test Plan:** ``` python test/quantization/test_qat.py -k test_qat_api_deprecation ``` Also manual testing: ``` 'IntXQuantizationAwareTrainingConfig' is deprecated and will be removed in a future release. Please use the following API instead: base_config = Int8DynamicActivationInt4WeightConfig(group_size=32) quantize_(model, QATConfig(base_config, step="prepare")) # train (not shown) quantize_(model, QATConfig(base_config, step="convert")) Alternatively, if you prefer to pass in fake quantization configs: activation_config = IntxFakeQuantizeConfig(torch.int8, "per_token", is_symmetric=False) weight_config = IntxFakeQuantizeConfig(torch.int4, group_size=32) qat_config = QATConfig( activation_config=activation_config, weight_config=weight_config, step="prepare", ) quantize_(model, qat_config) Please see #2630 for more details. IntXQuantizationAwareTrainingConfig(activation_config=None, weight_config=None) ``` ghstack-source-id: f6988d7 Pull Request resolved: #2641

**Summary:** Deprecates QAT APIs that should no longer be used. Print helpful deprecation warning to help users migrate. **Test Plan:** ``` python test/quantization/test_qat.py -k test_qat_api_deprecation ``` Also manual testing: ``` 'IntXQuantizationAwareTrainingConfig' is deprecated and will be removed in a future release. Please use the following API instead: base_config = Int8DynamicActivationInt4WeightConfig(group_size=32) quantize_(model, QATConfig(base_config, step="prepare")) # train (not shown) quantize_(model, QATConfig(base_config, step="convert")) Alternatively, if you prefer to pass in fake quantization configs: activation_config = IntxFakeQuantizeConfig(torch.int8, "per_token", is_symmetric=False) weight_config = IntxFakeQuantizeConfig(torch.int4, group_size=32) qat_config = QATConfig( activation_config=activation_config, weight_config=weight_config, step="prepare", ) quantize_(model, qat_config) Please see #2630 for more details. IntXQuantizationAwareTrainingConfig(activation_config=None, weight_config=None) ``` [ghstack-poisoned]

**Summary:** Deprecates QAT APIs that should no longer be used. Print helpful deprecation warning to help users migrate. **Test Plan:** ``` python test/quantization/test_qat.py -k test_qat_api_deprecation ``` Also manual testing: ``` 'IntXQuantizationAwareTrainingConfig' is deprecated and will be removed in a future release. Please use the following API instead: base_config = Int8DynamicActivationInt4WeightConfig(group_size=32) quantize_(model, QATConfig(base_config, step="prepare")) # train (not shown) quantize_(model, QATConfig(base_config, step="convert")) Alternatively, if you prefer to pass in fake quantization configs: activation_config = IntxFakeQuantizeConfig(torch.int8, "per_token", is_symmetric=False) weight_config = IntxFakeQuantizeConfig(torch.int4, group_size=32) qat_config = QATConfig( activation_config=activation_config, weight_config=weight_config, step="prepare", ) quantize_(model, qat_config) Please see #2630 for more details. IntXQuantizationAwareTrainingConfig(activation_config=None, weight_config=None) ``` ghstack-source-id: ac1b30e Pull Request resolved: #2641

**Summary:** Deprecates QAT APIs that should no longer be used. Print helpful deprecation warning to help users migrate. **Test Plan:** ``` python test/quantization/test_qat.py -k test_qat_api_deprecation ``` Also manual testing: ``` >>> from torchao.quantization.qat import IntXQuantizationAwareTrainingConfig >>> IntXQuantizationAwareTrainingConfig() 'IntXQuantizationAwareTrainingConfig' is deprecated and will be removed in a future release. Please use the following API instead: base_config = Int8DynamicActivationInt4WeightConfig(group_size=32) quantize_(model, QATConfig(base_config, step="prepare")) # train (not shown) quantize_(model, QATConfig(base_config, step="convert")) Alternatively, if you prefer to pass in fake quantization configs: activation_config = IntxFakeQuantizeConfig(torch.int8, "per_token", is_symmetric=False) weight_config = IntxFakeQuantizeConfig(torch.int4, group_size=32) qat_config = QATConfig( activation_config=activation_config, weight_config=weight_config, step="prepare", ) quantize_(model, qat_config) Please see #2630 for more details. IntXQuantizationAwareTrainingConfig(activation_config=None, weight_config=None) ``` [ghstack-poisoned]

**Summary:** Deprecates QAT APIs that should no longer be used. Print helpful deprecation warning to help users migrate. **Test Plan:** ``` python test/quantization/test_qat.py -k test_qat_api_deprecation ``` Also manual testing: ``` 'IntXQuantizationAwareTrainingConfig' is deprecated and will be removed in a future release. Please use the following API instead: base_config = Int8DynamicActivationInt4WeightConfig(group_size=32) quantize_(model, QATConfig(base_config, step="prepare")) # train (not shown) quantize_(model, QATConfig(base_config, step="convert")) Alternatively, if you prefer to pass in fake quantization configs: activation_config = IntxFakeQuantizeConfig(torch.int8, "per_token", is_symmetric=False) weight_config = IntxFakeQuantizeConfig(torch.int4, group_size=32) qat_config = QATConfig( activation_config=activation_config, weight_config=weight_config, step="prepare", ) quantize_(model, qat_config) Please see #2630 for more details. IntXQuantizationAwareTrainingConfig(activation_config=None, weight_config=None) ``` ghstack-source-id: bb3fc80 Pull Request resolved: #2641

**Summary:** Deprecates QAT APIs that should no longer be used. Print helpful deprecation warning to help users migrate. **Test Plan:** ``` python test/quantization/test_qat.py -k test_qat_api_deprecation ``` Also manual testing: ``` >>> from torchao.quantization.qat import IntXQuantizationAwareTrainingConfig >>> IntXQuantizationAwareTrainingConfig() 'IntXQuantizationAwareTrainingConfig' is deprecated and will be removed in a future release. Please use the following API instead: base_config = Int8DynamicActivationInt4WeightConfig(group_size=32) quantize_(model, QATConfig(base_config, step="prepare")) # train (not shown) quantize_(model, QATConfig(base_config, step="convert")) Alternatively, if you prefer to pass in fake quantization configs: activation_config = IntxFakeQuantizeConfig(torch.int8, "per_token", is_symmetric=False) weight_config = IntxFakeQuantizeConfig(torch.int4, group_size=32) qat_config = QATConfig( activation_config=activation_config, weight_config=weight_config, step="prepare", ) quantize_(model, qat_config) Please see #2630 for more details. IntXQuantizationAwareTrainingConfig(activation_config=None, weight_config=None) ``` [ghstack-poisoned]

**Summary:** Deprecates QAT APIs that should no longer be used. Print helpful deprecation warning to help users migrate. **Test Plan:** ``` python test/quantization/test_qat.py -k test_qat_api_deprecation ``` Also manual testing: ``` 'IntXQuantizationAwareTrainingConfig' is deprecated and will be removed in a future release. Please use the following API instead: base_config = Int8DynamicActivationInt4WeightConfig(group_size=32) quantize_(model, QATConfig(base_config, step="prepare")) # train (not shown) quantize_(model, QATConfig(base_config, step="convert")) Alternatively, if you prefer to pass in fake quantization configs: activation_config = IntxFakeQuantizeConfig(torch.int8, "per_token", is_symmetric=False) weight_config = IntxFakeQuantizeConfig(torch.int4, group_size=32) qat_config = QATConfig( activation_config=activation_config, weight_config=weight_config, step="prepare", ) quantize_(model, qat_config) Please see #2630 for more details. IntXQuantizationAwareTrainingConfig(activation_config=None, weight_config=None) ``` ghstack-source-id: 58738f1 Pull Request resolved: #2641

**Summary:** Deprecates QAT APIs that should no longer be used. Print helpful deprecation warning to help users migrate. **Test Plan:** ``` python test/quantization/test_qat.py -k test_qat_api_deprecation ``` Also manual testing: ``` >>> from torchao.quantization.qat import IntXQuantizationAwareTrainingConfig >>> IntXQuantizationAwareTrainingConfig() 'IntXQuantizationAwareTrainingConfig' is deprecated and will be removed in a future release. Please use the following API instead: base_config = Int8DynamicActivationInt4WeightConfig(group_size=32) quantize_(model, QATConfig(base_config, step="prepare")) # train (not shown) quantize_(model, QATConfig(base_config, step="convert")) Alternatively, if you prefer to pass in fake quantization configs: activation_config = IntxFakeQuantizeConfig(torch.int8, "per_token", is_symmetric=False) weight_config = IntxFakeQuantizeConfig(torch.int4, group_size=32) qat_config = QATConfig( activation_config=activation_config, weight_config=weight_config, step="prepare", ) quantize_(model, qat_config) Please see #2630 for more details. IntXQuantizationAwareTrainingConfig(activation_config=None, weight_config=None) ``` [ghstack-poisoned]

**Summary:** Deprecates QAT APIs that should no longer be used. Print helpful deprecation warning to help users migrate. **Test Plan:** ``` python test/quantization/test_qat.py -k test_qat_api_deprecation ``` Also manual testing: ``` 'IntXQuantizationAwareTrainingConfig' is deprecated and will be removed in a future release. Please use the following API instead: base_config = Int8DynamicActivationInt4WeightConfig(group_size=32) quantize_(model, QATConfig(base_config, step="prepare")) # train (not shown) quantize_(model, QATConfig(base_config, step="convert")) Alternatively, if you prefer to pass in fake quantization configs: activation_config = IntxFakeQuantizeConfig(torch.int8, "per_token", is_symmetric=False) weight_config = IntxFakeQuantizeConfig(torch.int4, group_size=32) qat_config = QATConfig( activation_config=activation_config, weight_config=weight_config, step="prepare", ) quantize_(model, qat_config) Please see #2630 for more details. IntXQuantizationAwareTrainingConfig(activation_config=None, weight_config=None) ``` ghstack-source-id: 7e0bda7 Pull Request resolved: #2641

**Summary:** Deprecates QAT APIs that should no longer be used. Print helpful deprecation warning to help users migrate. **Test Plan:** ``` python test/quantization/test_qat.py -k test_qat_api_deprecation ``` Also manual testing: ``` >>> from torchao.quantization.qat import IntXQuantizationAwareTrainingConfig >>> IntXQuantizationAwareTrainingConfig() 'IntXQuantizationAwareTrainingConfig' is deprecated and will be removed in a future release. Please use the following API instead: base_config = Int8DynamicActivationInt4WeightConfig(group_size=32) quantize_(model, QATConfig(base_config, step="prepare")) # train (not shown) quantize_(model, QATConfig(base_config, step="convert")) Alternatively, if you prefer to pass in fake quantization configs: activation_config = IntxFakeQuantizeConfig(torch.int8, "per_token", is_symmetric=False) weight_config = IntxFakeQuantizeConfig(torch.int4, group_size=32) qat_config = QATConfig( activation_config=activation_config, weight_config=weight_config, step="prepare", ) quantize_(model, qat_config) Please see #2630 for more details. IntXQuantizationAwareTrainingConfig(activation_config=None, weight_config=None) ``` [ghstack-poisoned]

**Summary:** Deprecates QAT APIs that should no longer be used. Print helpful deprecation warning to help users migrate. **Test Plan:** ``` python test/quantization/test_qat.py -k test_qat_api_deprecation ``` Also manual testing: ``` 'IntXQuantizationAwareTrainingConfig' is deprecated and will be removed in a future release. Please use the following API instead: base_config = Int8DynamicActivationInt4WeightConfig(group_size=32) quantize_(model, QATConfig(base_config, step="prepare")) # train (not shown) quantize_(model, QATConfig(base_config, step="convert")) Alternatively, if you prefer to pass in fake quantization configs: activation_config = IntxFakeQuantizeConfig(torch.int8, "per_token", is_symmetric=False) weight_config = IntxFakeQuantizeConfig(torch.int4, group_size=32) qat_config = QATConfig( activation_config=activation_config, weight_config=weight_config, step="prepare", ) quantize_(model, qat_config) Please see #2630 for more details. IntXQuantizationAwareTrainingConfig(activation_config=None, weight_config=None) ``` ghstack-source-id: ad4bd4e Pull Request resolved: #2641

**Summary:** Deprecates QAT APIs that should no longer be used. Print helpful deprecation warning to help users migrate. **Test Plan:** ``` python test/quantization/test_qat.py -k test_qat_api_deprecation ``` Also manual testing: ``` >>> from torchao.quantization.qat import IntXQuantizationAwareTrainingConfig >>> IntXQuantizationAwareTrainingConfig() 'IntXQuantizationAwareTrainingConfig' is deprecated and will be removed in a future release. Please use the following API instead: base_config = Int8DynamicActivationInt4WeightConfig(group_size=32) quantize_(model, QATConfig(base_config, step="prepare")) # train (not shown) quantize_(model, QATConfig(base_config, step="convert")) Alternatively, if you prefer to pass in fake quantization configs: activation_config = IntxFakeQuantizeConfig(torch.int8, "per_token", is_symmetric=False) weight_config = IntxFakeQuantizeConfig(torch.int4, group_size=32) qat_config = QATConfig( activation_config=activation_config, weight_config=weight_config, step="prepare", ) quantize_(model, qat_config) Please see #2630 for more details. IntXQuantizationAwareTrainingConfig(activation_config=None, weight_config=None) ``` [ghstack-poisoned]

**Summary:** Deprecates QAT APIs that should no longer be used. Print helpful deprecation warning to help users migrate. **Test Plan:** ``` python test/quantization/test_qat.py -k test_qat_api_deprecation ``` Also manual testing: ``` 'IntXQuantizationAwareTrainingConfig' is deprecated and will be removed in a future release. Please use the following API instead: base_config = Int8DynamicActivationInt4WeightConfig(group_size=32) quantize_(model, QATConfig(base_config, step="prepare")) # train (not shown) quantize_(model, QATConfig(base_config, step="convert")) Alternatively, if you prefer to pass in fake quantization configs: activation_config = IntxFakeQuantizeConfig(torch.int8, "per_token", is_symmetric=False) weight_config = IntxFakeQuantizeConfig(torch.int4, group_size=32) qat_config = QATConfig( activation_config=activation_config, weight_config=weight_config, step="prepare", ) quantize_(model, qat_config) Please see #2630 for more details. IntXQuantizationAwareTrainingConfig(activation_config=None, weight_config=None) ``` ghstack-source-id: 1961c9c Pull Request resolved: #2641

jerryzh168 · 2025-07-31T21:38:24Z

torchao/quantization/qat/utils.py

+    _LOGGED_DEPRECATED_CLASSES.add(old_api_object.__class__)
+    logger = logging.getLogger(old_api_object.__module__)
+    logger.warning(
+        """'%s' is deprecated and will be removed in a future release. Please use the following API instead:


is there a concrete time for deprecation?

yeah it's in the issue, deprecated this version (0.13.0) and removed the next (0.14.0)

jerryzh168 · 2025-07-31T21:44:46Z

torchao/quantization/qat/api.py

@@ -256,9 +257,13 @@ class IntXQuantizationAwareTrainingConfig(AOBaseConfig):
    activation_config: Optional[FakeQuantizeConfigBase] = None
    weight_config: Optional[FakeQuantizeConfigBase] = None

+    def __post_init__(self):


I heard typing_extensions.deprecated is more IDE friendly: pytorch/pytorch#153892 (comment)

hmm I don't think torchao has a dependency on typing_extensions? Do we want to add a new one for this?

I see, OK probably not worth to add deps since it looks like right now there is no deps for torchao

jerryzh168 · 2025-07-31T23:08:03Z

torchao/quantization/qat/utils.py

+
+
+# log deprecation warning only once per class
+_LOGGED_DEPRECATED_CLASSES = set[type]()


actually I'm also looking at how to do warning once, does warnings.warn work for you?

the global set approach is the only way to do it

it seems that warnings.warn by default only prints once though? https://docs.python.org/3/library/warnings.html#the-warnings-filter

Looks like I was able to get warnings.warn to log only once, updated

**Summary:** Deprecates QAT APIs that should no longer be used. Print helpful deprecation warning to help users migrate. **Test Plan:** ``` python test/quantization/test_qat.py -k test_qat_api_deprecation ``` Also manual testing: ``` >>> from torchao.quantization.qat import IntXQuantizationAwareTrainingConfig >>> IntXQuantizationAwareTrainingConfig() 'IntXQuantizationAwareTrainingConfig' is deprecated and will be removed in a future release. Please use the following API instead: base_config = Int8DynamicActivationInt4WeightConfig(group_size=32) quantize_(model, QATConfig(base_config, step="prepare")) # train (not shown) quantize_(model, QATConfig(base_config, step="convert")) Alternatively, if you prefer to pass in fake quantization configs: activation_config = IntxFakeQuantizeConfig(torch.int8, "per_token", is_symmetric=False) weight_config = IntxFakeQuantizeConfig(torch.int4, group_size=32) qat_config = QATConfig( activation_config=activation_config, weight_config=weight_config, step="prepare", ) quantize_(model, qat_config) Please see #2630 for more details. IntXQuantizationAwareTrainingConfig(activation_config=None, weight_config=None) ``` [ghstack-poisoned]

**Summary:** Deprecates QAT APIs that should no longer be used. Print helpful deprecation warning to help users migrate. **Test Plan:** ``` python test/quantization/test_qat.py -k test_qat_api_deprecation ``` Also manual testing: ``` 'IntXQuantizationAwareTrainingConfig' is deprecated and will be removed in a future release. Please use the following API instead: base_config = Int8DynamicActivationInt4WeightConfig(group_size=32) quantize_(model, QATConfig(base_config, step="prepare")) # train (not shown) quantize_(model, QATConfig(base_config, step="convert")) Alternatively, if you prefer to pass in fake quantization configs: activation_config = IntxFakeQuantizeConfig(torch.int8, "per_token", is_symmetric=False) weight_config = IntxFakeQuantizeConfig(torch.int4, group_size=32) qat_config = QATConfig( activation_config=activation_config, weight_config=weight_config, step="prepare", ) quantize_(model, qat_config) Please see #2630 for more details. IntXQuantizationAwareTrainingConfig(activation_config=None, weight_config=None) ``` ghstack-source-id: 7ac9f3b Pull Request resolved: #2641

**Summary:** Deprecates QAT APIs that should no longer be used. Print helpful deprecation warning to help users migrate. **Test Plan:** ``` python test/quantization/test_qat.py -k test_qat_api_deprecation ``` Also manual testing: ``` >>> from torchao.quantization.qat import IntXQuantizationAwareTrainingConfig >>> IntXQuantizationAwareTrainingConfig() 'IntXQuantizationAwareTrainingConfig' is deprecated and will be removed in a future release. Please use the following API instead: base_config = Int8DynamicActivationInt4WeightConfig(group_size=32) quantize_(model, QATConfig(base_config, step="prepare")) # train (not shown) quantize_(model, QATConfig(base_config, step="convert")) Alternatively, if you prefer to pass in fake quantization configs: activation_config = IntxFakeQuantizeConfig(torch.int8, "per_token", is_symmetric=False) weight_config = IntxFakeQuantizeConfig(torch.int4, group_size=32) qat_config = QATConfig( activation_config=activation_config, weight_config=weight_config, step="prepare", ) quantize_(model, qat_config) Please see #2630 for more details. IntXQuantizationAwareTrainingConfig(activation_config=None, weight_config=None) ``` [ghstack-poisoned]

**Summary:** Deprecates QAT APIs that should no longer be used. Print helpful deprecation warning to help users migrate. **Test Plan:** ``` python test/quantization/test_qat.py -k test_qat_api_deprecation ``` Also manual testing: ``` 'IntXQuantizationAwareTrainingConfig' is deprecated and will be removed in a future release. Please use the following API instead: base_config = Int8DynamicActivationInt4WeightConfig(group_size=32) quantize_(model, QATConfig(base_config, step="prepare")) # train (not shown) quantize_(model, QATConfig(base_config, step="convert")) Alternatively, if you prefer to pass in fake quantization configs: activation_config = IntxFakeQuantizeConfig(torch.int8, "per_token", is_symmetric=False) weight_config = IntxFakeQuantizeConfig(torch.int4, group_size=32) qat_config = QATConfig( activation_config=activation_config, weight_config=weight_config, step="prepare", ) quantize_(model, qat_config) Please see #2630 for more details. IntXQuantizationAwareTrainingConfig(activation_config=None, weight_config=None) ``` ghstack-source-id: 33a6f38 Pull Request resolved: #2641

andrewor14 added 5 commits July 29, 2025 13:02

Deprecate old QAT APIs

1e88ebf

**Summary:** Deprecates QAT APIs that should no longer be used. Print helpful deprecation warning to help users migrate. **Test Plan:** ``` python test/quantization/test_qat.py -k test_qat_api_deprecation ``` [ghstack-poisoned]

This was referenced Jul 30, 2025

[bc-breaking] Generalize FakeQuantizeConfig beyond intx #2628

Merged

New multi-step QAT API #2629

Merged

meta-cla bot added the CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. label Jul 30, 2025

andrewor14 added the topic: improvement Use this tag if this PR is an improvement (doesn't fit into any of the other categories) label Jul 30, 2025

andrewor14 requested review from vkuzo, jerryzh168 and drisspg July 30, 2025 22:55

andrewor14 added 2 commits July 30, 2025 15:58

Update base for Update on "Deprecate old QAT APIs"

12e8c3f

**Summary:** Deprecates QAT APIs that should no longer be used. Print helpful deprecation warning to help users migrate. **Test Plan:** ``` python test/quantization/test_qat.py -k test_qat_api_deprecation ``` [ghstack-poisoned]

Update on "Deprecate old QAT APIs"

2ed5e50

**Summary:** Deprecates QAT APIs that should no longer be used. Print helpful deprecation warning to help users migrate. **Test Plan:** ``` python test/quantization/test_qat.py -k test_qat_api_deprecation ``` [ghstack-poisoned]

andrewor14 added 2 commits July 30, 2025 16:00

andrewor14 added 2 commits July 30, 2025 16:02

andrewor14 added 2 commits July 30, 2025 16:10

andrewor14 added 2 commits July 31, 2025 06:35

andrewor14 added 2 commits July 31, 2025 06:41

andrewor14 added 2 commits July 31, 2025 14:06

andrewor14 added 2 commits July 31, 2025 14:19

jerryzh168 reviewed Jul 31, 2025

View reviewed changes

jerryzh168 approved these changes Jul 31, 2025

View reviewed changes

jerryzh168 reviewed Jul 31, 2025

View reviewed changes

andrewor14 added 2 commits August 1, 2025 08:14

andrewor14 added 2 commits August 1, 2025 08:43

andrewor14 mentioned this pull request Aug 1, 2025

Add NVFP4 QAT #2666

Open

andrewor14 changed the base branch from gh/andrewor14/15/base to main August 4, 2025 16:59

Merge branch 'main' into gh/andrewor14/15/head

3f06429

andrewor14 merged commit eb3f4aa into main Aug 5, 2025
33 of 38 checks passed

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

Deprecate old QAT APIs #2641

Deprecate old QAT APIs #2641

Uh oh!

andrewor14 commented Jul 30, 2025 •

edited

Loading

Uh oh!

pytorch-bot bot commented Jul 30, 2025 •

edited

Loading

Uh oh!

jerryzh168 Jul 31, 2025

Uh oh!

andrewor14 Jul 31, 2025

Uh oh!

jerryzh168 Jul 31, 2025

Uh oh!

andrewor14 Jul 31, 2025

Uh oh!

jerryzh168 Jul 31, 2025

Uh oh!

jerryzh168 Jul 31, 2025

Uh oh!

drisspg Jul 31, 2025

Uh oh!

jerryzh168 Jul 31, 2025

Uh oh!

andrewor14 Aug 1, 2025

Uh oh!

Uh oh!

Uh oh!



		# log deprecation warning only once per class
		_LOGGED_DEPRECATED_CLASSES = set[type]()

Deprecate old QAT APIs #2641

Deprecate old QAT APIs #2641

Uh oh!

Conversation

andrewor14 commented Jul 30, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

pytorch-bot bot commented Jul 30, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/ao/2641

✅ You can merge normally! (2 Unrelated Failures)

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

andrewor14 commented Jul 30, 2025 •

edited

Loading

pytorch-bot bot commented Jul 30, 2025 •

edited

Loading