[PyTorch][torch.compile] Make quantizers opaque value objects by pggPL · Pull Request #3152 · NVIDIA/TransformerEngine

pggPL · 2026-06-29T09:36:33Z

Description

Tensorless quantizers in TE (MXFP8, FP8 blockwise, FP8 current-scaling, NVFP4)
are fully described by a handful of plain, reproducible scalars — they hold no
live tensors and no process groups. This PR turns them into opaque value
objects so torch.compile can treat them as baked-in constants: two
quantizers with the same configuration become interchangeable, hashable, and
reconstructible inside an FX graph.

Quantizers that hold live state (delayed-scaling Float8Quantizer, which keeps
scale/amax tensors) and any user-defined quantizer keep the default
identity semantics, so the change is opt-in and backward compatible. On older
PyTorch builds without the opaque-object API the registration is a graceful
no-op.

Along the way this also un-breaks the existing test_torch_compile.py suite:
that file lived on main but was never wired into CI, and its
test_autocast_nested_custom case (nested te.autocast with multiple
CustomRecipe instances) was failing because of the CustomRecipe state-caching
bug fixed here. The file is now run in CI and passes.

Type of change

Documentation change (change only to the documentation, either a fix or a new content)
Bug fix (non-breaking change which fixes an issue)
New feature (non-breaking change which adds functionality)
Breaking change (fix or feature that would cause existing functionality to not work as expected)
Infra/Build change
Code refactoring

Changes

Add opt-in value-object identity to the base Quantizer
(_value_fields / _value_key / __eq__ / __hash__). Returning None
from _value_fields() (the default) keeps identity semantics.
New module transformer_engine/pytorch/dynamo.py holding the
torch.compile glue: __fx_repr__, value-key reconstruction and
register_value_opaque_quantizer (gracefully no-op without PyTorch's
opaque-object API).
Register MXFP8Quantizer, Float8BlockQuantizer,
Float8CurrentScalingQuantizer and NVFP4Quantizer as value opaque types
(the deprecated amax_reduction_group is never part of the value).
Fix CustomRecipe state caching in TransformerEngineBaseModule.set_meta_tensor:
rebuild quantizers when the CustomRecipe instance changes (e.g. nested
te.autocast regions) instead of reusing the first recipe's state, since
every CustomRecipe shares the CustomRecipeState type but carries its own
qfactory. This fixes the previously-failing test_autocast_nested_custom.
Enable tests/pytorch/test_torch_compile.py in the L0_pytorch_unittest QA
suite (it existed on main but was never run in CI), and add the quantizer
value-object tests to it. Bringing it into CI required fixing the existing
CustomRecipe torch.compile path: the qfactory now dispatches on
QuantizerRole.tensor_type supplied by ToyLinear.get_quantizer_roles.
Guard the value-object path against a stored amax reduction group: __fx_repr__
already rejects any quantizer holding a process group, and __eq__ / __hash__
now raise too. The group is excluded from the value key, so a stored group would
otherwise compare/hash equal to a groupless quantizer and let torch.compile
reuse a graph that skips the reduction. Pass the group per quantize call instead.

Checklist:

I have read and followed the contributing guidelines
The functionality is complete
I have commented my code, particularly in hard-to-understand areas
I have made corresponding changes to the documentation
My changes generate no new warnings
I have added tests that prove my fix is effective or that my feature works
New and existing unit tests pass locally with my changes

…ompile Give tensorless quantizers (MXFP8, FP8 blockwise, FP8 current-scaling, NVFP4) value-object semantics so torch.compile can treat them as baked-in constants: - Add opt-in value identity to the base Quantizer (_value_fields / _value_key / __eq__ / __hash__). Quantizers holding live tensors (delayed-scaling Float8Quantizer) and custom quantizers keep identity semantics. - New transformer_engine/pytorch/dynamo.py houses the torch.compile glue: __fx_repr__, value-key reconstruction and register_value_opaque_quantizer (gracefully a no-op on PyTorch builds without the opaque-object API). - Register the four tensorless quantizers as value opaque types. Also fix CustomRecipe state caching in TransformerEngineBaseModule: set_meta_tensor now rebuilds quantizers when the CustomRecipe instance changes (e.g. nested te.autocast regions) instead of reusing the first recipe's state, since every CustomRecipe shares the CustomRecipeState type but carries its own qfactory. Move the quantizer value-object tests into tests/pytorch/test_torch_compile.py and add that file to the L0 pytorch unittest QA suite. Signed-off-by: Pawel Gadzinski <pgadzinski@nvidia.com>

…globals Follow-up to the value-opaque quantizer support: - Remove the module-level _QUANTIZER_VALUE_REGISTRY (qualname -> class) and _quantizer_from_value_key. __fx_repr__ now captures the quantizer class directly in the FX globals and reconstructs via _rebuild_quantizer(cls, items), matching how PyTorch's own value opaque types (e.g. DTensor placements) reconstruct themselves. This removes global mutable state and the qualname collision risk. - Consolidate the quantizer value-object tests in test_torch_compile.py down to two functions and exercise reconstruction through the public __fx_repr__ path instead of internal helpers. Signed-off-by: Pawel Gadzinski <pgadzinski@nvidia.com>

Replace the single dynamo.py module with a dynamo/ package so the torch.compile glue can grow with a clear responsibility split across the stacked branches. This branch owns the value-opaque quantizer layer. * dynamo/quantizer_opaque.py -- register_value_opaque_quantizer and helpers * dynamo/__init__.py -- re-exports the public API so callers keep importing from transformer_engine.pytorch.dynamo unchanged Signed-off-by: Pawel Gadzinski <pgadzinski@nvidia.com>

A value-opaque quantizer must not carry live distributed state. Scan the quantizer attributes in __fx_repr__ and raise TypeError if any holds a torch.distributed.ProcessGroup (e.g. a non-None deprecated amax_reduction_group), so it cannot be silently baked into a torch.compile FX graph. Clarify the related comments accordingly. Signed-off-by: Pawel Gadzinski <pgadzinski@nvidia.com>

NVFP4Quantizer is registered as a value-opaque quantizer but was missing from the value-semantics / __fx_repr__ round-trip test. Add it to _VALUE_QUANTIZERS (skipped without CUDA, which it needs to construct). Signed-off-by: Pawel Gadzinski <pgadzinski@nvidia.com>

…__/__hash__ The amax reduction group is excluded from the value key, so a value quantizer that stored one would compare/hash equal to a groupless one and let torch.compile reuse a graph that skips the reduction. __eq__/__hash__ now raise (mirroring __fx_repr__, which already rejects any process-group-bearing quantizer). The group should be passed per quantize call, not stored on the quantizer. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> Signed-off-by: Pawel Gadzinski <pgadzinski@nvidia.com>

Add is_value_opaque_quantizer() + the _te_compile_value_opaque flag stamped at registration, so dynamo-traced code can detect registered quantizers (and fall back to eager for unregistered ones). Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> Signed-off-by: Pawel Gadzinski <pgadzinski@nvidia.com>

…fp4 value key - Narrow register_opaque_type except to (RuntimeError, TypeError): the API is already imported above, so ImportError/AttributeError there only mask real errors. - Add test_quantizer_value_object_fullgraph exercising torch.compile(fullgraph=True) end-to-end to verify opaque-type registration took effect. - Restore missing NVFP4Quantizer._with_random_sign_mask assignment required by _value_fields()/_value_key(). Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> Signed-off-by: Pawel Gadzinski <pgadzinski@nvidia.com>

pggPL · 2026-06-29T09:38:39Z

/te-ci pytorch

greptile-apps · 2026-06-29T09:42:30Z

Greptile Summary

This PR turns stateless TE quantizers (MXFP8Quantizer, Float8BlockQuantizer, Float8CurrentScalingQuantizer, NVFP4Quantizer) into value objects so torch.compile can treat them as baked-in FX-graph constants — adding __eq__/__hash__ by configuration key, __fx_repr__ for codegen, and opaque-type registration via a new dynamo module. It also fixes a CustomRecipe state-caching bug and wires test_torch_compile.py into CI.

Value-object infrastructure (quantized_tensor.py, dynamo/quantizer_opaque.py): _value_fields() / _value_key() / _BASE_VALUE_FIELDS on the base Quantizer, plus _rebuild_quantizer and graceful opaque-type registration that no-ops on older PyTorch builds.
Per-quantizer registration (four tensor files): each adds _value_fields() and a module-level register_value_opaque_quantizer(...) call; NVFP4Quantizer additionally stores _with_random_sign_mask and exposes a _rebuild_derived_state hook to reconstruct the non-hashable rht_matrix tensor after FX reconstruction.
Test coverage (test_torch_compile.py): new test_quantizer_value_object (repr round-trip + hash), test_value_quantizer_rejects_process_group, and test_quantizer_value_object_fullgraph (opaque type inside a fullgraph=True compile).

Confidence Score: 4/5

Safe to merge after fixing the copy() omission in NVFP4Quantizer; all other paths are correct.

The copy() method in NVFP4Quantizer was not updated to pass with_random_sign_mask to the constructor, leaving _with_random_sign_mask always True in any copy. This makes q != q.copy() when the original was constructed with with_random_sign_mask=False, and if the copy ever reaches the FX-graph reconstruction path, _rebuild_derived_state would call get_rht_matrix with the wrong flag and silently produce an incorrect rht_matrix. The rest of the PR is well-designed and correct.

transformer_engine/pytorch/tensor/nvfp4_tensor.py — the copy() method needs with_random_sign_mask=self._with_random_sign_mask added to its constructor call.

Important Files Changed

Filename	Overview
transformer_engine/pytorch/dynamo/quantizer_opaque.py	New module implementing FX repr / opaque-type registration; `_rebuild_quantizer` correctly handles base fields and the `_rebuild_derived_state` hook; registration is gracefully no-op on old PyTorch.
transformer_engine/pytorch/quantized_tensor.py	Adds `_value_fields`, `_value_key`, `__eq__`, `__hash__` to `Quantizer`; base attrs match `Quantizer.__init__` exactly; process-group guard is applied consistently.
transformer_engine/pytorch/tensor/nvfp4_tensor.py	Adds `_with_random_sign_mask` storage, `_rebuild_derived_state` hook, and `_value_fields`; `copy()` omits `with_random_sign_mask` from the constructor call, leaving `_with_random_sign_mask` incorrect in the copy.
transformer_engine/pytorch/tensor/float8_blockwise_tensor.py	Adds `_value_fields` returning the five scalar config attributes; straightforward and complete.
transformer_engine/pytorch/tensor/float8_tensor.py	Adds `_value_fields` for `Float8CurrentScalingQuantizer` with correct exclusion of `amax_reduction_group`.
transformer_engine/pytorch/tensor/mxfp8_tensor.py	Minimal `_value_fields` returning `("dtype",)` plus module-level registration; correct.
tests/pytorch/test_torch_compile.py	Adds value-object tests covering round-trip repr, hash, equality, process-group rejection, and full-graph compile; tests are well-structured and cover edge cases.
transformer_engine/pytorch/dynamo/init.py	Simple module init re-exporting the two public symbols; clean.

Sequence Diagram

%%{init: {'theme': 'neutral'}}%%
sequenceDiagram
    participant User
    participant Compile as torch.compile
    participant Q as Quantizer
    participant RQ as _rebuild_quantizer
    participant NQ as NVFP4Quantizer

    User->>Compile: "compile(fn, fullgraph=True)"
    Compile->>Q: __fx_repr__()
    Q->>Q: _value_key() checks no process group
    Q-->>Compile: repr string + globals dict
    Note over Compile: bake repr into FX graph constant
    Compile->>RQ: cls.__new__(cls) then set fields from items
    RQ->>NQ: _rebuild_derived_state()
    NQ->>NQ: get_rht_matrix(_with_random_sign_mask, device)
    RQ-->>Compile: reconstructed quantizer
    Compile-->>User: compiled fn with quantizer baked in

%%{init: {'theme': 'base', 'themeVariables': {"darkMode": true, "background": "#0d1117", "primaryColor": "#21262d", "primaryTextColor": "#e6edf3", "primaryBorderColor": "#8b949e", "lineColor": "#8b949e", "textColor": "#e6edf3", "edgeLabelBackground": "#161b22", "actorBkg": "#21262d", "actorBorder": "#8b949e", "actorTextColor": "#e6edf3", "actorLineColor": "#8b949e", "signalColor": "#8b949e", "signalTextColor": "#e6edf3", "noteBkgColor": "#373320", "noteBorderColor": "#d4a72c", "noteTextColor": "#f0e6c0", "labelBoxBkgColor": "#21262d", "labelBoxBorderColor": "#8b949e", "labelTextColor": "#e6edf3", "loopTextColor": "#e6edf3", "activationBkgColor": "#30363d", "activationBorderColor": "#8b949e"}}}%%
sequenceDiagram
    participant User
    participant Compile as torch.compile
    participant Q as Quantizer
    participant RQ as _rebuild_quantizer
    participant NQ as NVFP4Quantizer

    User->>Compile: "compile(fn, fullgraph=True)"
    Compile->>Q: __fx_repr__()
    Q->>Q: _value_key() checks no process group
    Q-->>Compile: repr string + globals dict
    Note over Compile: bake repr into FX graph constant
    Compile->>RQ: cls.__new__(cls) then set fields from items
    RQ->>NQ: _rebuild_derived_state()
    NQ->>NQ: get_rht_matrix(_with_random_sign_mask, device)
    RQ-->>Compile: reconstructed quantizer
    Compile-->>User: compiled fn with quantizer baked in

Comments Outside Diff (1)

transformer_engine/pytorch/tensor/nvfp4_tensor.py, line 231-245 (link)

copy() doesn't propagate _with_random_sign_mask to the new constructor, so the copy's _with_random_sign_mask is always True (the default), regardless of the original's value. This breaks value-key equality (q != q.copy() when with_random_sign_mask=False) and, more critically, poisons FX-graph reconstruction: _rebuild_derived_state reads self._with_random_sign_mask to call get_rht_matrix, so a reconstructed quantizer would silently compute the wrong rht_matrix for any quantizer that was originally created with with_random_sign_mask=False.

_{Reviews (6): Last reviewed commit: "Drop trivial value-equality boilerplate ..." | Re-trigger Greptile}

…trip _rebuild_quantizer only restores value-key fields, so a reconstructed NVFP4Quantizer was missing the derived rht_matrix tensor (not hashable, so not in the value key) and failed at copy()/quantize time. Add a _rebuild_derived_state hook (called by _rebuild_quantizer) that NVFP4Quantizer uses to rebuild rht_matrix from _with_random_sign_mask (lru_cache -> cheap). Extend test_quantizer_value_object to also quantize with the original and the rebuilt quantizer and require bit-exact results (gated on HW support), so a field the kernel needs but the value key omits can no longer slip through. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> Signed-off-by: Pawel Gadzinski <pgadzinski@nvidia.com>

kshitij12345

Overall LGTM, would be good to resolve the inline comments before merging.

Move the ProcessGroup guard out of the (overridable) __fx_repr__ into Quantizer._value_key -- the single point every value-materialization path (__eq__/__hash__/__fx_repr__) goes through -- so a custom __fx_repr__ can no longer bypass it. Generalizes the old amax-only check to any field holding a ProcessGroup. Add a test that a value quantizer carrying a live group raises. Addresses review on NVIDIA#3152. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> Signed-off-by: Pawel Gadzinski <pgadzinski@nvidia.com>

…assthrough Replace the trivial pass-through fullgraph test with one that drives each production quantizer through a minimal custom op (quantize + dequantize) under torch.compile(fullgraph=True) and compares to eager -- so the opaque-type registration is actually exercised inside the graph (a graph break would make fullgraph=True raise). Op registration sits right before the test. Also drop stale comments referencing the old __fx_repr__-side process-group guard. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> Signed-off-by: Pawel Gadzinski <pgadzinski@nvidia.com>

…paque flag - rht_matrix_random_sign_mask_t is a device-independent int derived from _with_random_sign_mask (the device only places a throwaway tensor); fix the misleading comment. - Explain why registration uses a class attribute, not a registry set: is_value_opaque_quantizer is traced inside the compile graph and dynamo can bake a getattr constant but cannot do 'type(q) in set' on the opaque class. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> Signed-off-by: Pawel Gadzinski <pgadzinski@nvidia.com>

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> Signed-off-by: Pawel Gadzinski <pgadzinski@nvidia.com>

is_opaque_value_type(cls) sat between the import guard and the register_opaque_type guard, so on a partial/experimental opaque-object build it could raise RuntimeError/TypeError and crash TE import. Move it inside the same except so the 'registration never crashes import' promise holds for both calls. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> Signed-off-by: Pawel Gadzinski <pgadzinski@nvidia.com>

pggPL · 2026-06-29T14:35:58Z

/te-ci pytorch

ptrendx · 2026-06-30T15:19:44Z

+    a, b = factory(), factory()
+    # Same config -> equal, same hash, interchangeable as a dict/set key.
+    assert a is not b
+    assert a == b
+    assert hash(a) == hash(b)
+    assert {a: "x"}[b] == "x"
+    # Different config -> not equal.
+    assert a != factory(**other_kwargs)


What exactly are we testing in this part?

nothing, I can remove it it seems to be claude boilerplate

ptrendx · 2026-06-30T15:23:32Z

+def test_value_quantizer_rejects_process_group():
+    """A value quantizer holding a live ProcessGroup must refuse to be turned
+    into a value key / FX constant (raise), not silently drop the group."""
+    import torch.distributed as dist  # pylint: disable=import-outside-toplevel
+
+    created = not dist.is_initialized()
+    if created:
+        dist.init_process_group(backend="gloo", store=dist.HashStore(), rank=0, world_size=1)
+    try:
+        q = MXFP8Quantizer(fp8_dtype=tex.DType.kFloat8E4M3)
+        q.amax_reduction_group = dist.group.WORLD
+        # Every value-materialization path must reject it (hash, eq, __fx_repr__).
+        with pytest.raises(TypeError):
+            hash(q)
+        with pytest.raises(TypeError):
+            q.__fx_repr__()
+    finally:
+        if created:
+            dist.destroy_process_group()


I'm not sure I understand the purpose of this test (and the logic needed to make it work). Is it here to make sure that if somebody accidentally adds the group to the quantizer then it would loudly fail e.g. in the CI?

It was suggestion of @kshitij12345

Context around the test - There was a possible code-path previously (now fixed) where having a ProcessGroup on the quantizer wouldn't trigger the unsupported TypeError (and fail later). This test verifies that the error would be raised with a clear message.

ptrendx · 2026-06-30T20:40:44Z

+    # Bypass ``__init__`` and restore the value attributes directly: the value
+    # items already capture every value-defining field (including derived ones),
+    # and the constructors have heterogeneous signatures / side effects.


Huh? What are the issues with just calling the constructor?

Many things have different names in constructor than as parameters or are not present as args (like ".internal" or ".optimize_for_gemm")

ptrendx · 2026-06-30T20:45:20Z

+    # ``register_opaque_type`` requires ``__fx_repr__`` to already exist on the
+    # class, so attach it before registering.
+    if "__fx_repr__" not in cls.__dict__:
+        cls.__fx_repr__ = _quantizer_fx_repr


Then why aren'twe just adding that to the Quantizer class? I feel that I miss some context here as a lot of this file feels like a workaround to some issues that are not described.

We can do this, I wanted to have as much of torch.compile specific logic here

ptrendx · 2026-06-30T20:47:15Z

+        """
+        return None
+
+    def _check_value_has_no_process_group(self) -> None:


How often is this method called? Only once during torch.compile or does it need to be called every time during the graph launch to check the guards?

every time - I make it faster by just checking amax_reduction_group

ptrendx · 2026-06-30T20:59:20Z

        return quantizer

+    def _value_fields(self) -> Tuple[str, ...]:
+        return ("dtype", "block_len", "amax_epsilon", "force_pow_2_scales", "block_scaling_dim")


There are some ways of getting that list without hardcoding it again, e.g. using inspect.get_annotations in the cases like we have where we provide all of the data members up front or via vars.

I see It's quite messy, but it is result of mess in quantizers. There are some fields called in the constructor (and arguments names sometimes does not match the field name). Some fields are set like .internal or .block_len are not arguments to constructor. And some of them are excluded on purpose like amax_reduction_group or rht_matrix.

ptrendx · 2026-06-30T21:01:55Z

+        ``_rebuild_quantizer`` calls this hook to rebuild it; the ``lru_cache`` on
+        :func:`get_rht_matrix` makes an already-seen (flag, device) a cheap hit.
+        """
+        self.rht_matrix = get_rht_matrix(self._with_random_sign_mask, torch.cuda.current_device())


Using current device here is a little annoying since in principle it does not have to be the right device (which instead should be the input device).

It's not problem of this PR, we use current device in constructor.

Move the _VALUE_OPAQUE_FLAG setattr to the end of register_value_opaque_quantizer, after register_opaque_type succeeds (or the type is already opaque). Previously the flag was set up front, so is_value_opaque_quantizer reported True even when the opaque-object API was missing or registration raised, since both paths are swallowed. Eager value semantics (__eq__/__hash__/__fx_repr__) are independent of the flag, so this only tightens the predicate to mean torch actually knows the type as opaque. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> Signed-off-by: Pawel Gadzinski <pgadzinski@nvidia.com>

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> Signed-off-by: Pawel Gadzinski <pgadzinski@nvidia.com>

_check_value_has_no_process_group ran on every guard eval (via __eq__/__hash__) and scanned all of vars(self) recursively. The only attribute that can hold a ProcessGroup is the deprecated amax_reduction_group, so check it directly (O(1)) and drop the _contains_process_group helper. Same guarantee, off the hot path. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> Signed-off-by: Pawel Gadzinski <pgadzinski@nvidia.com>

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> Signed-off-by: Pawel Gadzinski <pgadzinski@nvidia.com>

# Conflicts: # transformer_engine/pytorch/tensor/float8_blockwise_tensor.py # transformer_engine/pytorch/tensor/float8_tensor.py # transformer_engine/pytorch/tensor/mxfp8_tensor.py # transformer_engine/pytorch/tensor/nvfp4_tensor.py

Remove the a==b / hash / dict-key block that just exercised Python's own dict semantics; equality and hashing are still covered by the __fx_repr__ round-trip (rebuilt == a, hash match) and the bit-exact kernel check. other_kwargs is now unused, so drop it from the parametrization and both test signatures. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> Signed-off-by: Pawel Gadzinski <pgadzinski@nvidia.com>

pggPL and others added 8 commits June 29, 2026 11:25

pggPL requested a review from ksivaman as a code owner June 29, 2026 09:36

pggPL mentioned this pull request Jun 29, 2026

[PyTorch][torch.compile] Add TensorProto mechanism #3153

Open

13 tasks

greptile-apps Bot reviewed Jun 29, 2026

View reviewed changes

Comment thread transformer_engine/pytorch/tensor/nvfp4_tensor.py

Comment thread transformer_engine/pytorch/quantized_tensor.py

kshitij12345 approved these changes Jun 29, 2026

View reviewed changes

Comment thread transformer_engine/pytorch/dynamo/quantizer_opaque.py

Comment thread transformer_engine/pytorch/dynamo/quantizer_opaque.py Outdated

pggPL and others added 2 commits June 29, 2026 14:46

greptile-apps Bot reviewed Jun 29, 2026

View reviewed changes

Comment thread transformer_engine/pytorch/dynamo/quantizer_opaque.py Outdated

pggPL and others added 3 commits June 29, 2026 15:29

Reword opaque-flag comment: self-contained, no Linear reference

2c3c5df

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> Signed-off-by: Pawel Gadzinski <pgadzinski@nvidia.com>

ptrendx reviewed Jun 30, 2026

View reviewed changes

pggPL and others added 6 commits June 30, 2026 23:44

Drop verbose comments around value-opaque flag stamping

9db604f

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> Signed-off-by: Pawel Gadzinski <pgadzinski@nvidia.com>

Shorten amax_reduction_group check comment

fe5e5db

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> Signed-off-by: Pawel Gadzinski <pgadzinski@nvidia.com>

Uh oh!

Conversation

pggPL commented Jun 29, 2026

Description

Type of change

Changes

Checklist:

Uh oh!

pggPL commented Jun 29, 2026

Uh oh!

greptile-apps Bot commented Jun 29, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Greptile Summary

Confidence Score: 4/5

Important Files Changed

Sequence Diagram

Comments Outside Diff (1)

Uh oh!

Uh oh!

Uh oh!

kshitij12345 left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

pggPL commented Jun 29, 2026

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!

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!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

greptile-apps Bot commented Jun 29, 2026 •

edited

Loading