ml-explore Release Notes

4 Einträge aus 2 Quellen. Zuletzt aktualisiert: 29. September 2026

Feed abonnieren:RSS API

29. September 2026

Datum aus der Quelle.

Erstmals gesehen am 6. Oktober 2026.

Mlx 0.32.3: Fehlerbehebungen, correction-Parameter für std und var

Mlx v0.32.3 behebt mehrere Fehler (u. a. Scan und Sort bei Achsen der Größe null, einen Deadlock durch mx.clear_streams(), Integer-pow mit negativem Exponenten und pad mit negativen Achsen), ergänzt einen correction-Parameter für std und var, macht die Achsenangabe bei put_along_axis optional, setzt bei take_along_axis den Standardwert -1 und propagiert NaN in Arg-Reduktionen.

What's Changed

Fix scan and sort for a zero-size axis by @eyupcanakman in https://github.com/ml-explore/mlx/pull/4340
Add a correction parameter in std and var by @prady0t in https://github.com/ml-explore/mlx/pull/4348
Fix tests/run.py stuck in macOS CI by @zcbenz in https://github.com/ml-explore/mlx/pull/4410
Extension Example Updated by @jagrit06 in https://github.com/ml-explore/mlx/pull/4412
Fix sorted gather_qmm NAX row overflow above 32K by @PhilipJohnBasile in https://github.com/ml-explore/mlx/pull/3922
Fix deadlock caused by mx.clear_streams() holding GIL by @aleroot in https://github.com/ml-explore/mlx/pull/4413
Fix integer pow zeroing a whole SIMD vector on a negative exponent by @ayaangazali in https://github.com/ml-explore/mlx/pull/4354
Make concurrency cap on load adaptative so that I/O scales with the machine by @aleroot in https://github.com/ml-explore/mlx/pull/4408
Fix pad with an axes subset and negative axes by @kapellirohith in https://github.com/ml-explore/mlx/pull/4364
Use sdpa_vector_2pass_1_gqa for GQA size 12 and 16 by @dudududukim in https://github.com/ml-explore/mlx/pull/4380
python: Make axis of put_along_axis optional by @ayaangazali in https://github.com/ml-explore/mlx/pull/4360
python: Make axis default to -1 in take_along_axis by @aaishwarymishra in https://github.com/ml-explore/mlx/pull/4368
Propagate NaN in arg reductions by @atirna in https://github.com/ml-explore/mlx/pull/4291 …

Mehr anzeigenWeniger anzeigen

Originalquelle(öffnet in neuem Tab)Problem melden

25. August 2026

Datum aus der Quelle.

Erstmals gesehen am 6. Oktober 2026.

Mlx von ml-explore

Mlx 0.32.2: Zahlreiche Fehlerbehebungen und Option force_fused

Mlx v0.32.2 behebt zahlreiche Fehler (u. a. Cast subnormaler Floats zu bool, divmod bei Floats, median mit NaN, fft-Transformationen über Achsenteilmengen), ergänzt die Option force_fused für scaled_dot_product_attention, validiert eps und Optimizer-Betas und härtet das Lesen von GGUF-Metadaten.

What's Changed

Preserve subnormal float values when casting to bool by @reckylurker in https://github.com/ml-explore/mlx/pull/4224
Support assigning through a bare Ellipsis index by @Adityaj0 in https://github.com/ml-explore/mlx/pull/4314
Fix divmod truncating the quotient for floats by @ayaangazali in https://github.com/ml-explore/mlx/pull/4108
Add force_fused option to scaled_dot_product_attention by @hojin12312 in https://github.com/ml-explore/mlx/pull/4185
Reject a negative eps in the normalization layers by @ayaangazali in https://github.com/ml-explore/mlx/pull/4312
Bound GGUF metadata string/array values against the file mapping by @x14ngch3n in https://github.com/ml-explore/mlx/pull/4212
Read each K/V byte once in gqa-8 decode attention by @dudududukim in https://github.com/ml-explore/mlx/pull/4077
Fix fft vmap and jvp for transforms over a subset of axes by @kapellirohith in https://github.com/ml-explore/mlx/pull/4138
Fix median dropping NaN by @devteamaegis in https://github.com/ml-explore/mlx/pull/4146
Fix the CPU scan over a size one axis with a padded stride by @kapellirohith in https://github.com/ml-explore/mlx/pull/4139
Validate the optimizer betas at construction by @ayaangazali in https://github.com/ml-explore/mlx/pull/4310
RMSNormVJP backward writes a full {n_rows, D} gw_temp intermediate by @JasonHonKL in https://github.com/ml-explore/mlx/pull/4293 …

Mehr anzeigenWeniger anzeigen

Originalquelle(öffnet in neuem Tab)Problem melden

18. August 2026

Datum aus der Quelle.

Erstmals gesehen am 6. Oktober 2026.

Mlx von ml-explore

Mlx 0.32.1: Fehlerbehebungen, schnelleres RMSNorm unter CUDA

Mlx v0.32.1 enthält Fehlerbehebungen (u. a. GGUF-Metadaten-Arrays mit int64, zufälliger Zustand in compile), Verbesserungen der Dokumentation, eine schnellere RMSNorm-Vorwärtsberechnung unter CUDA und zero-copy-Import von Host-Buffern über mx.array(host_buffer, copy=False).

What's Changed

Fix int64 type cast error when loading GGUF metadata arrays by @danlee2002 in https://github.com/ml-explore/mlx/pull/3823
Document default values in normalization layer docstrings by @Pablosinyores in https://github.com/ml-explore/mlx/pull/3819
Warn at configure time when NAX kernels are disabled by @pierre427 in https://github.com/ml-explore/mlx/pull/3824
[CUDA][Improvement] RMSNorm forward speed up by @nastya236 in https://github.com/ml-explore/mlx/pull/3850
Fix captured random state in compile by @angeloskath in https://github.com/ml-explore/mlx/pull/3828
Fix JIT preamble header filter matching project paths containing "Xcode" by @apocryphx in https://github.com/ml-explore/mlx/pull/3873
Document default value of p in dropout layer docstrings by @ayaangazali in https://github.com/ml-explore/mlx/pull/3870
[WIP] [CUDA] fsdp by @nastya236 in https://github.com/ml-explore/mlx/pull/3768
Fix triplet_loss docstring to document the reduced output shape by @ayaangazali in https://github.com/ml-explore/mlx/pull/3884
Zero-copy CPU import: mx.array(host_buffer, copy=False) on unified memory by @HaoXuAI in https://github.com/ml-explore/mlx/pull/3872
Round MLX_SDPA_BLOCKS up to a multiple of 32 by @pierre427 in https://github.com/ml-explore/mlx/pull/3875
Reuse Metal WAR tracking hash tables by @neilmehta24 in https://github.com/ml-explore/mlx/pull/3882 …

Mehr anzeigenWeniger anzeigen

Originalquelle(öffnet in neuem Tab)Problem melden

7. Juli 2026

Datum aus der Quelle.

Erstmals gesehen am 6. Oktober 2026.

Mlx von ml-explore

Mlx 0.32.0: Determinante in linalg, JACCL-Barrier und CUDA-Fixes

Mlx v0.32.0 bringt unter anderem Determinante und Sign-Log-Determinante in mlx.core.linalg, eine Barrier für JACCL, klarere Fehlermeldungen bei Int32-Überlauf von Dimensionen, device_count() gibt ohne GPU 0 zurück sowie mehrere CUDA- und Indexierungs-Fehlerbehebungen.

What's Changed

Generate qmm implementaions with cmake by @zcbenz in https://github.com/ml-explore/mlx/pull/3424
Enable swap for all CI building CUDA by @zcbenz in https://github.com/ml-explore/mlx/pull/3437
Bump minor by @angeloskath in https://github.com/ml-explore/mlx/pull/3438
[CUDA] Fix qmm_naive K-tail dispatch for FP quantized kernels by @Lyxot in https://github.com/ml-explore/mlx/pull/3445
Keep gguflib input-validation asserts active in release builds by @qflen in https://github.com/ml-explore/mlx/pull/3436
Reuse nightly build's ccache for release by @zcbenz in https://github.com/ml-explore/mlx/pull/3458
Add barrier to JACCL by @Isalia20 in https://github.com/ml-explore/mlx/pull/3459
Add determinant and sign-log-determinant functions to mlx.core.linalg by @abhilashreddys in https://github.com/ml-explore/mlx/pull/3416
Define ST_F8_E8M0 by @pcuenca in https://github.com/ml-explore/mlx/pull/3448
Clearer error when shape dimension overflows int32 by @serenposh in https://github.com/ml-explore/mlx/pull/3425
[CUDA] Fix half type matmul in cutlass kernels by @zcbenz in https://github.com/ml-explore/mlx/pull/3469
Fix indexing bug in slice update with op by @angeloskath in https://github.com/ml-explore/mlx/pull/3483
Make device_count() return 0 when there is no GPU by @zcbenz in https://github.com/ml-explore/mlx/pull/3486
Compute contiguity from the actual occupied data by @sofinvalery in https://github.com/ml-explore/mlx/pull/3475 …

Mehr anzeigenWeniger anzeigen

Originalquelle(öffnet in neuem Tab)Problem melden

ml-explore Release Notes

Mlx 0.32.3: Fehlerbehebungen, correction-Parameter für std und var

What's Changed

Mlx 0.32.2: Zahlreiche Fehlerbehebungen und Option force_fused

What's Changed

Mlx 0.32.1: Fehlerbehebungen, schnelleres RMSNorm unter CUDA

What's Changed

Mlx 0.32.0: Determinante in linalg, JACCL-Barrier und CUDA-Fixes

What's Changed

Ähnlich wie ml-explore, mit neuen Einträgen