80bcd1c658
[CUDA] Fix half type matmul in cutlass kernels ( #3469 )
2026-05-06 08:35:53 +09:00
fa4320d5fa
[CUDA] Handle residue k in qmm_naive ( #3379 )
2026-04-18 13:30:07 +09:00
d142de6a20
[CUDA] gather_mm ( #3414 )
2026-04-17 16:53:44 +09:00
2ffafe07f4
[CUDA] 3/5/6-bit quants for qmm_naive ( #3352 )
2026-04-01 20:13:01 +09:00
1c9ee2f655
[CUDA] Fallback QMM ( #3315 )
2026-04-01 12:41:26 +09:00
0ff1115a46
[CUDA] Implement BlockMaskedMM ( #3299 )
...
Co-authored-by: Cheng <git@zcbenz.com >
2026-03-27 06:57:26 +09:00
0bdbfdb838
[CUDA] Implement MaskedScatter ( #3151 )
2026-03-15 10:33:55 +09:00
5d1700493a
[CUDA] Add FFT support ( #3243 )
2026-03-14 21:02:19 +09:00
a9573f92f6
[CUDA] Implement SegmentedMM ( #3238 )
2026-03-11 13:31:43 -07:00
9d03a1b0d9
[CUDA] Support 3/5/6-bit quants in QMV ( #3236 )
2026-03-10 19:09:48 +09:00
be872ebdef
[CUDA] implement Hadamard transform ( #3179 )
2026-03-05 09:34:19 +01:00
90e38f7b93
Fix qmv_impl for small N ( #3096 )
2026-02-05 17:33:36 -08:00
4912cc47c2
Fp qmv ( #2984 )
2026-01-27 06:33:06 -08:00
099dcc0f4c
Expose to/from fp8 in Python and don't auto-convert fp8 when loading from safetensors ( #2985 )
2026-01-13 15:48:21 -08:00
1d21d0e696
[CUDA] Implement gather_mm_rhs ( #2902 )
2025-12-24 09:42:56 +09:00
1eef1d155c
Metal/CPU nvfp4 and mxfp8 ( #2946 )
2025-12-22 20:45:19 -08:00
b3825ac149
Add Masked Scatter ( #2663 )
...
Co-authored-by: Awni Hannun <awni@apple.com >
Co-authored-by: Angelos Katharopoulos <katharas@gmail.com >
Co-authored-by: Angelos Katharopoulos <a_katharopoulos@apple.com >
2025-11-19 14:53:32 -08:00
70560b6bd5
Add mode parameter for quantization ( #2499 )
...
* add mode parameter for quantization
* mxfp4 quantize/dequantize + start of optional biases
* mxfp4 works
* speedup
* cpu mxfp4
* fix
* fix test tol
* fix
* refactor
* add quant mode enum
2025-08-28 06:45:26 -07:00
4822c3dbe9
[CUDA] Implement DynamicSlice/DynamicSliceUpdate ( #2533 )
...
* Move DynamicSlice to gpu/primitives
* Implement compute_dynamic_offset in CUDA
2025-08-26 07:31:39 +09:00
f4c8888cbe
[CUDA] Fix stride of singleton dims before passing to cuDNN ( #2521 )
2025-08-21 08:55:26 +09:00
ac85ddfdb7
[CUDA] Add GEMM-based fallback convolution kernels ( #2511 )
...
* Add gemm_conv
* Add gemm_grouped_conv
2025-08-20 10:06:22 +09:00
1ba18ff7d9
[CUDA] Fix conv grads with groups ( #2495 )
...
* Put reshape utils in one file
* [CUDA] Fix conv grads with groups
* Put the reshape utils in gpu/copy.h
2025-08-16 10:09:18 +09:00
86c6a15571
[CUDA] Backward convolution ( #2431 )
2025-08-01 09:54:05 +09:00
8b25ce62d5
Add tests for export including control flow models and quantized models ( #2430 )
...
* Add tests for export, including control flow export and quantized model export.
* Skip quantization related test for CUDA backend.
2025-07-31 11:06:26 -07:00
6f5874a2f2
[CUDA] Initial implementation of Convolution with cuDNN ( #2385 )
...
* Link with cuDNN
* Initial implementation
* Remove backend apis
* Fix recording cudnn conv
* More unused backend apis
* Fix C++ conv tests
* include cudnn as python dep
* Install libcudnn9-dev-cuda-12 in CI
* cudnn only accepts contiguous inputs
* Switch to backend apis
* Plan needs to be kept alive
* Turn off tf32
* Add cache
* Test the native cuda graph api
* Set cudnn stream before execution
* Make LRUCache more like a normal container
* Do error check for cublas handle
* Zero-initilizing array
* Use tf32 for conv
* Skip TestConv.test_torch_conv_2D test
---------
Co-authored-by: Awni Hannun <awni@apple.com >
2025-07-25 08:12:10 +09:00
d7734edd9f
fix complex reduce + nan propagation in min and max ( #2377 )
2025-07-15 18:19:47 -07:00
e7d2ebadd2
[CUDA] Affine quantize ( #2354 )
...
* affine quantize and dequantize kernels
* format
* fix
* format
2025-07-14 15:45:44 -07:00
8347575ba1
[CUDA] Implement Scan kernel ( #2347 )
...
* Contiguous scan
* Strided scan
* Enable tests
* Fix failing logaddexp test
* Use cexpf in Metal
2025-07-10 16:54:12 -07:00
8b9a3f3cea
Align mlx::core::max op nan propagation with NumPy ( #2339 )
...
* Make max op NaN propagation rules align with numpy
* Adding benchmarks and testing for max op nanpropagation
* Pre-commit formatting
* Fix max complex64 nan propagation and add test
* Improve the cpp unittest
* Only check nans on non-integral types in simd_reduce_impl.
* Cleanup using namespace alias
* Add cpu Max nanpropagation. Fix a small fib in cpu max dispatch data types for int8/int16.
* Make the max nanpropagation test more meaningful for integer types
* Remove tuple unpacking syntax to comply with earlier python versions. Add cuda skip to nanpropagation tests, fix cuda implementation in a separate PR.
2025-07-09 11:26:27 -07:00
4a9b29a875
MoE backward improvements ( #2335 )
2025-07-07 17:59:53 -07:00
dd4f53db63
use fp32 for testing, add more complex ops ( #2322 )
2025-07-01 07:30:00 -07:00
772f471ff2
[CUDA] Fix reductions ( #2314 )
2025-06-27 12:59:20 -07:00
cad5c0241c
[CUDA] synch properly waits for all tasks to finish and clear ( #2303 )
...
* cuda synch properly waits for all tasks to finish and clear
* fix copy
2025-06-17 12:03:25 -07:00
b8022c578a
divmod, partition, sort fixes ( #2302 )
2025-06-16 18:49:32 -07:00
bc53f8293f
Cuda bug fixes 2 ( #2298 )
...
* more bug fixes
* more bug fixes
* format
2025-06-16 13:14:46 -07:00
c552ff2451
[CUDA] Fix back-end bugs and enable corresponding tests ( #2296 )
...
* Fix some cuda back-end bugs and enable corresponding tests
* more fixes
* enable more tests
* format
2025-06-16 08:45:40 -07:00
4fda5fbdf9
add python testing for cuda with ability to skip list of tests ( #2295 )
2025-06-15 10:56:48 -07:00
Cheng