GeezerVison

Branch selector

Commit history

1. Commits on Nov 12, 2026

   • Update XLA dependency to use revision http://github.com/openxla/xla/commit/0bc40b0190fb743b0a300d1541e512a933f8e357

     Show description for 12612cf
     Google-ML-Automation

     committed
     12612cfView commit details

     Copy full SHA for 12612cf

     Browse repository at this point

   • sharding-in-types + reduction_order, plus cleanups

     Show description for 1a1e875
     

     mattjj

     authored and

     Google-ML-Automation

     committed
     1a1e875View commit details

     Copy full SHA for 1a1e875

     Browse repository at this point
2. Commits on Nov 11, 2026

   • Enable jax_accurate_erf_inv by default

     Show description for 1fc0c01
     

     hawkinsp

     authored and

     Google-ML-Automation

     committed
     1fc0c01View commit details

     Copy full SHA for 1fc0c01

     Browse repository at this point

   • [CI] Record test source location and unparameterized name in pytest JUnit XML.

     Show description for a32a9aa
     

     belitskiy

     authored and

     Google-ML-Automation

     committed
     a32a9aaView commit details

     Copy full SHA for a32a9aa

     Browse repository at this point

   • Merge pull request #42463 from robinbishtt:fix-kron-vander-bool

     Show description for debd1db
     Google-ML-Automation

     committed
     debd1dbView commit details

     Copy full SHA for debd1db

     Browse repository at this point

   • Enable more accurate default erfc in JAX.

     Show description for f886230
     

     hawkinsp

     authored and

     Google-ML-Automation

     committed
     f886230View commit details

     Copy full SHA for f886230

     Browse repository at this point

   • Return signed ULP differences from `ulp_diff` and make `ulp_diff_mpmath` private.

     Show description for d43d446
     

     hawkinsp

     authored and

     Google-ML-Automation

     committed
     d43d446View commit details

     Copy full SHA for d43d446

     Browse repository at this point

   • Fix jnp.kron and jnp.vander for boolean inputs

     Show description for 22e530b
     robinbishtt

     committed
     22e530bView commit details

     Copy full SHA for 22e530b

     Browse repository at this point

   • Merge pull request #42467 from zazabap:dct-rfft

     Show description for 3db9797
     Google-ML-Automation

     committed
     3db9797View commit details

     Copy full SHA for 3db9797

     Browse repository at this point

   • Merge pull request #42478 from tanveer-arch:fix-choose-narrow-dtype-large-n

     Show description for 02c1926
     Google-ML-Automation

     committed
     02c1926View commit details

     Copy full SHA for 02c1926

     Browse repository at this point

   • Expose `shared_ifrt_loaded_executable()` on `PyMpmdLoadedExecutable`

     Show description for 6ddaa25
     Google-ML-Automation

     committed
     6ddaa25View commit details

     Copy full SHA for 6ddaa25

     Browse repository at this point

   • Remove deprecated keyword arguments from `jnp.clip` type signature.

     Show description for af02062
     

     Jake VanderPlas

     authored and

     Google-ML-Automation

     committed
     af02062View commit details

     Copy full SHA for af02062

     Browse repository at this point

   • [jax-buildcop] Eliminate redundant XLA compilations in `hyp1f1`, `hyp2f1`, and `betaln` gradient tests

     Show description for 9f8a2ea
     Google-ML-Automation

     committed
     9f8a2eaView commit details

     Copy full SHA for 9f8a2ea

     Browse repository at this point

   • Merge pull request #42497 from jakevdp:bessel-doc

     Show description for 941f3e8
     Google-ML-Automation

     committed
     941f3e8View commit details

     Copy full SHA for 941f3e8

     Browse repository at this point

   • [jax-buildcop] Use `libstdc++` with `linux_glibc_2_35` sysroot under `--config=tsan` in `.bazelrc`

     Show description for bbd7366
     Google-ML-Automation

     committed
     bbd7366View commit details

     Copy full SHA for bbd7366

     Browse repository at this point

   • [jax-buildcop] Enable `TF_FORCE_GPU_ALLOW_GROWTH` in `jax2tf` GPU tests

     Show description for e352c12
     Google-ML-Automation

     committed
     e352c12View commit details

     Copy full SHA for e352c12

     Browse repository at this point

   • [doc] improve Bessel function docs & cross-referencing

     jakevdp

     committed
     7bda3abView commit details

     Copy full SHA for 7bda3ab

     Browse repository at this point

   • [jax-buildcop] Suppress spurious bfloat16 isinf warning on Windows in nan_to_num test.

     Show description for 946c7a1
     Google-ML-Automation

     committed
     946c7a1View commit details

     Copy full SHA for 946c7a1

     Browse repository at this point

   • Handle unknown arg_names when explaining tracing cache misses.

     Show description for d972a82
     Google-ML-Automation

     committed
     d972a82View commit details

     Copy full SHA for d972a82

     Browse repository at this point

   • [jax-buildcop] Reuse compiled shapes in Bessel kn tests to avoid shard timeouts

     Show description for 37da041
     Google-ML-Automation

     committed
     37da041View commit details

     Copy full SHA for 37da041

     Browse repository at this point

   • Use call_sharding_rule in SearchSorted

     Show description for c1e242b
     

     yashk2810

     authored and

     Google-ML-Automation

     committed
     c1e242bView commit details

     Copy full SHA for c1e242b

     Browse repository at this point

   • Allow mixed `Auto` meshes in `SearchSorted` and `jnp.histogram`.

     Show description for 87648ea
     

     levskaya

     authored and

     Google-ML-Automation

     committed
     87648eaView commit details

     Copy full SHA for 87648ea

     Browse repository at this point

   • [Mosaic GPU][NFC] Remove the default arguments of `SmemDescriptor.slice_i64`.

     Show description for c08d31c
     

     allanrenucci

     authored and

     Google-ML-Automation

     committed
     c08d31cView commit details

     Copy full SHA for c08d31c

     Browse repository at this point

   • Rename the jax_shard_map_partitioning configuration flag to jax_partitioning.

     Show description for b82be55
     

     yashk2810

     authored and

     Google-ML-Automation

     committed
     b82be55View commit details

     Copy full SHA for b82be55

     Browse repository at this point

   • Fix jnp.choose for narrow index dtypes and large choice counts

     Show description for 91a6f34
     tanveer-arch

     committed
     91a6f34View commit details

     Copy full SHA for 91a6f34

     Browse repository at this point

   • [Mosaic GPU] Allow allocating up to 576 TMEM columns on sm_107f.

     Show description for 447f46e
     

     golechwierowicz

     authored and

     Google-ML-Automation

     committed
     447f46eView commit details

     Copy full SHA for 447f46e

     Browse repository at this point

   • Align TMEM transfer layout inference with TMEMRef.load/store.

     Show description for 1ad1de7
     

     golechwierowicz

     authored and

     Google-ML-Automation

     committed
     1ad1de7View commit details

     Copy full SHA for 1ad1de7

     Browse repository at this point

   • Plumb the target architecture into layout inference derivation rules.

     Show description for d430af3
     

     golechwierowicz

     authored and

     Google-ML-Automation

     committed
     d430af3View commit details

     Copy full SHA for d430af3

     Browse repository at this point

   • [jax-buildcop] Stop worker subprocess cleanly before joining in `ProfilerTest.testSubprocessProfiling`

     Show description for 6dca3ac
     Google-ML-Automation

     committed
     6dca3acView commit details

     Copy full SHA for 6dca3ac

     Browse repository at this point

   • [Mosaic GPU] Lift restriction on 128-lane A scales in sparse block-scaled MMA on sm_107f

     Show description for f5981d9
     

     golechwierowicz

     authored and

     Google-ML-Automation

     committed
     f5981d9View commit details

     Copy full SHA for f5981d9

     Browse repository at this point

   • Lift restriction on 128-lane A scales in collective M=256 block-scaled MMA on sm_107f.

     Show description for f719a9a
     

     golechwierowicz

     authored and

     Google-ML-Automation

     committed
     f719a9aView commit details

     Copy full SHA for f719a9a

     Browse repository at this point

   • jax.scipy.fft.dct: use rfft for real input

     Show description for f64b120
     zazabap

     committed
     f64b120View commit details

     Copy full SHA for f64b120

     Browse repository at this point

   • Update XLA dependency to use revision http://github.com/openxla/xla/commit/c308dbb884239531b82276135a785c1a04683ebb

     Show description for fd48df9
     Google-ML-Automation

     committed
     fd48df9View commit details

     Copy full SHA for fd48df9

     Browse repository at this point

   • [Mosaic:GPU] Parse collective attributes only once and store it in CustomCallResources during instantiation.

     Show description for 776c7a4
     

     PatriosTheGreat

     authored and

     Google-ML-Automation

     committed
     776c7a4View commit details

     Copy full SHA for 776c7a4

     Browse repository at this point

   • Reverts 1e384f92d5ce5a419a7c4ba4645cff88a7080c74

     Show description for 4473bda
     Google-ML-Automation

     committed
     4473bdaView commit details

     Copy full SHA for 4473bda

     Browse repository at this point