- 05 Jun, 2025 1 commit
-
-
Kirthi Shankar Sivamani authored
* Fix NVTE_FRAMEWORK=all Signed-off-by:
Kirthi Shankar Sivamani <ksivamani@nvidia.com> * fix Signed-off-by:
-
- 05 May, 2025 1 commit
-
-
jberchtold-nvidia authored
* Enforce input sharding of norm primitive does not shard hidden dim Signed-off-by:
Jeremy Berchtold <jberchtold@nvidia.com> * Fix partitioning issue in dact primitive causing NaN and add better shape checks before calling TE API Signed-off-by:
-
- 24 Apr, 2025 1 commit
-
-
jberchtold-nvidia authored
Introduce nvte_memset to provide a fill kernel that is faster than cudaMemsetAsync for small sizes (#1716) * nvte_memset fills single float value Signed-off-by:
-
- 22 Apr, 2025 1 commit
-
-
jberchtold-nvidia authored
* [JAX-Q] Single GPU current scaling for JAX Signed-off-by:
-
- 09 Apr, 2025 1 commit
-
-
Phuong Nguyen authored
* scaling enum abstract * rm NVTE_ from ScalingMode names * rework scaling mode enum in grouped gemm * fix norm sharding --------- Signed-off-by:Phuong Nguyen <phuonguyen@nvidia.com>
-
- 04 Apr, 2025 1 commit
-
-
Phuong Nguyen authored
* rename QuantizeAxis to QuantizeLayout, get_layout to get_data_layout, q_axis to q_layout * add fatten_axis option * added gated act to test encoder * sharding constraint fixes * fix padding when flattening first dim needs to be padded * update test sizes so that padding is tested * rm output sharding as it can be done in the flax module * sharding scale_inv for mxfp8 --------- Signed-off-by:
-
- 01 Apr, 2025 1 commit
-
-
Phuong Nguyen authored
* refactor + mxfp8 * added grouped gemm * rename linear to dense * added cublas init phase for groupedGemm * relax the tol of test encoder multiprocessing mxfp8 by 0.001 Signed-off-by:
Hua Huang <huah@nvidia.com> Co-authored-by:
-
- 07 Feb, 2025 1 commit
-
-
Przemek Tredak authored
Signed-off-by:Przemek Tredak <ptredak@nvidia.com>
-
- 02 Jan, 2025 1 commit
-
-
Kirthi Shankar Sivamani authored
Signed-off-by:
-
- 13 Nov, 2024 1 commit
-
-
Jennifer Zhou authored
fix an int conversion error Signed-off-by:Jennifer Zhou <jennifer@jezh.me>
-
- 06 Nov, 2024 1 commit
-
-
Hua Huang authored
* FFI for some transpose & activation functions Signed-off-by:
Phuong Nguyen <36155692+phu0ngng@users.noreply.github.com> Signed-off-by:
Hua Huang <huangh1994@outlook.com> --------- Signed-off-by:
pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> Co-authored-by:
-
- 31 Oct, 2024 1 commit
-
-
Phuong Nguyen authored
* lowering a dict of attrs * improve err message with line and func info * implement a product() for ffi dimensions --------- Signed-off-by:
-
- 24 Oct, 2024 1 commit
-
-
Hua Huang authored
[JAX] XLA Custom Calls with FFI for FusedAttnFwd, Quantize, Transpose, ActLuFP8, LayerNormForwardFP8FFI, and LayerNormBackwardFFI (#1263) * Add TransposeFFI, test passed Signed-off-by:
-
- 17 Oct, 2024 1 commit
-
-
Phuong Nguyen authored
* register CmdBufferCompatible traits via C++ API * renamed FFI_Traits * use register_ffi_target() --------- Signed-off-by:
-
- 14 Aug, 2024 1 commit
-
-
Phuong Nguyen authored
* implemented custom call with ffi in csrc * moved headers of misc to misc.h, add ffi.h * ActLu and DActLu lowering with ffi_lowering * CastTranspose with ffi_lowering * enabled cudaGraph * added 4d input test case to TestActivationLu * added operand_output_aliases for CastTranspose * added env var NVTE_JAX_WITH_FFI, default value = 1 * replace casting ActivationEnum by taking its value --------- Signed-off-by:
-
- 25 Jul, 2024 1 commit
-
-
Kirthi Shankar Sivamani authored
* Specify python version Signed-off-by:
-
- 17 Jun, 2024 1 commit
-
-
Alp Dener authored
replaced plain C asserts with NVTE_CHECK to avoid unused-variable warnings Signed-off-by:Alp Dener <adener@nvidia.com>
-
- 14 Jun, 2024 1 commit
-
-
Kirthi Shankar Sivamani authored
* Apply formatting Signed-off-by:
-
- 08 Jun, 2024 1 commit
-
-
Phuong Nguyen authored
* categorized `csrc/modules.cpp` Signed-off-by:
-
