Repository navigation
Fix nax condition for iphone - #3083
Merged
Merged
Conversation
Member
Author
|
Also bumped |
davidkoski
approved these changes
Jan 29, 2026
davidkoski
left a comment
Member
There was a problem hiding this comment.
Tested on iPhone 16 Pro and it now behaves as expected
davidkoski
approved these changes
Jan 29, 2026
davidkoski
left a comment
Member
There was a problem hiding this comment.
Tested on iPhone 16 Pro and it now behaves as expected
|
FYI @objectivecosta |
|
Oh wow, I wish I had seen this before! |
benjaminsoo
added a commit
to benjaminsoo/mlx-swift
that referenced
this pull request
Mar 2, 2026
iPhone 16 (A18, gen 17) was incorrectly detected as having Neural
Accelerator support, causing garbage output. Require gen >= 18 for
phone ('p') architectures.
Vendored mlx and mlx-c submodules as regular files to include the fix.
whothisbeme
added a commit
to whothisbeme/kokoro-ios
that referenced
this pull request
Mar 8, 2026
mlx-swift 0.30.2 produces static/corrupted audio on iPhone 16 Pro due to two NAX (Neural Accelerator) bugs: 1. Incorrect NAX hardware detection on A18 chip (ml-explore/mlx#3083) - A18 (gen 17) was wrongly detected as having NAX support (gen 18+) - Caused silently wrong numerical output instead of a crash 2. NAX overflow in ConvTransposed1d (ml-explore/mlx#3092) - Incorrect output for large tensors (time dim > ~8000) - Directly affects Kokoro's vocoder transposed convolutions Both fixes are included in mlx-swift 0.30.6. Using `from:` instead of `exact:` allows downstream consumers to resolve compatible versions. Relates to: mlalma/KokoroTestApp#7, ml-explore/mlx-swift#344
dakshaymehta
added a commit
to dakshaymehta/kokoro-ios
that referenced
this pull request
Apr 10, 2026
mlx-swift 0.30.2 has two NAX bugs that cause static/corrupted audio: - Incorrect NAX hardware detection (ml-explore/mlx#3083) - NAX overflow in ConvTransposed1d (ml-explore/mlx#3092) Both fixed in mlx-swift 0.30.6. Also changes pin style from exact to from to allow downstream consumers to resolve compatible versions. Ref: mlalma#29
This was referenced May 27, 2026
Jerry0523
added a commit
to Jerry0523/kokoro-ios
that referenced
this pull request
Sep 6, 2026
Update mlx-swift from exact 0.30.2 to from 0.30.6 mlx-swift 0.30.2 produces static/corrupted audio on iPhone 16 Pro due to two NAX (Neural Accelerator) bugs: 1. Incorrect NAX hardware detection on A18 chip (ml-explore/mlx#3083) - A18 (gen 17) was wrongly detected as having NAX support (gen 18+) - Caused silently wrong numerical output instead of a crash 2. NAX overflow in ConvTransposed1d (ml-explore/mlx#3092) - Incorrect output for large tensors (time dim > ~8000) - Directly affects Kokoro's vocoder transposed convolutions Both fixes are included in mlx-swift 0.30.6. Using `from:` instead of `exact:` allows downstream consumers to resolve compatible versions.
This was referenced Sep 11, 2026
linyubupa
added a commit
to Edge0-AI/Edge0
that referenced
this pull request
Sep 11, 2026
mlx <= 0.30.4 enables its NAX (tensor-core) matmul kernels for phone-class GPUs (`can_use_nax &= get_architecture_gen() >= 17`), which A18 / A18 Pro report as gen 17 even though that path returns silently wrong numbers: checkpoints load, LoRA/prerouter install, generation runs to completion, and the text comes out as incoherent mixed-language noise. Fixed upstream by ml-explore/mlx#3083 ("Fix nax condition for iphone") and ml-explore/mlx#3092 ("Fix for NAX overflow"), both released in 0.30.5 (see issue #8). * bump mlx / mlx-metal to 0.30.6 (mlx-lm stays at 0.31.0) * stay below 0.31.2: it makes default streams thread-local and the streaming expert loader builds mx arrays on pool threads, which then raises "There is no Stream(gpu, N) in current thread" * README / README_zh: the symptom and the one-line runtime upgrade Verified on the M4 Pro bench machine (applegpu_g16s, which never takes the NAX path): greedy token ids for both tiers are identical between 0.30.4 and 0.30.6 (16/16 per tier), `pytest` 60 passed / 1 skipped, `pytest -m slow` 2 passed. The A18-side effect still needs confirmation from the reporter.
pull Bot
pushed a commit
to vishalbelsare/Edge0
that referenced
this pull request
Sep 11, 2026
mlx <= 0.30.4 enables its NAX (tensor-core) matmul kernels for phone-class GPUs (`can_use_nax &= get_architecture_gen() >= 17`), which A18 / A18 Pro report as gen 17 even though that path returns silently wrong numbers: checkpoints load, LoRA/prerouter install, generation runs to completion, and the text comes out as incoherent mixed-language noise. Fixed upstream by ml-explore/mlx#3083 ("Fix nax condition for iphone") and ml-explore/mlx#3092 ("Fix for NAX overflow"), both released in 0.30.5 (see issue #8). * bump mlx / mlx-metal to 0.30.6 (mlx-lm stays at 0.31.0) * stay below 0.31.2: it makes default streams thread-local and the streaming expert loader builds mx arrays on pool threads, which then raises "There is no Stream(gpu, N) in current thread" * README / README_zh: the symptom and the one-line runtime upgrade Verified on the M4 Pro bench machine (applegpu_g16s, which never takes the NAX path): greedy token ids for both tiers are identical between 0.30.4 and 0.30.6 (16/16 per tier), `pytest` 60 passed / 1 skipped, `pytest -m slow` 2 passed. The A18-side effect still needs confirmation from the reporter.
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
5 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.
As title. Will eventually close ml-explore/mlx-swift-examples#462
CC @davidkoski