-
Notifications
You must be signed in to change notification settings - Fork 846
Pull requests: NVIDIA/cutlass
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
Make integer_subbyte Fully Compliant with is_integral
#1632
opened Jul 15, 2024 by
osayamenja
Loading…
Add Ampere GEMM example using Cute and CUTLASS 3.x
#1604
opened Jun 27, 2024 by
aacostadiaz
Loading…
Add GEMM Kernel Example for Hopper H100 Tensor Cores
#1578
opened Jun 7, 2024 by
IonThruster
Loading…
Update gemm_f16n_f16t_f32t_tensor_op_f32_sm80.cu with include "cutlas…
inactive-30d
#1569
opened Jun 3, 2024 by
houqi
Loading…
Allow scalar broadcasting in VisitorRowBroadcast and VisitorColBroadcast
feature request
New feature or request
#1539
opened May 16, 2024 by
tlrmchlsmth
Loading…
Fix template parameter
IterationsUnroll
type from int to bool
inactive-30d
#1534
opened May 11, 2024 by
peakcrosser7
Loading…
Update half.h - typo at line 138(unnecessary space before '1')
inactive-30d
#1527
opened May 8, 2024 by
sjbae1999
Loading…
add publication: ‘EVT: Accelerating Deep Learning Training with Epilo…
inactive-30d
#1526
opened May 7, 2024 by
reed-lau
Loading…
support data type u2 used in cutlass_library
inactive-30d
#1517
opened Apr 30, 2024 by
gavinchen430
Loading…
feat: support kFactor 8 used in mma tensor op tile iterator
inactive-30d
#1512
opened Apr 29, 2024 by
gavinchen430
Loading…
Add missing #include <memory> for definition of std::addressof.
inactive-30d
#1470
opened Apr 10, 2024 by
Gregory-Meyer
Loading…
Refactor to use FastDivmod for predicated strided dgrad iterators.
inactive-30d
#1453
opened Apr 3, 2024 by
ZelboK
Loading…
Add support for mixed 4-bit/8-bit data types GEMM
#1413
opened Mar 19, 2024 by
alexsamardzic
Loading…
Add couple configs into generator.py for mixed input MM
#1350
opened Feb 16, 2024 by
alexsamardzic
Loading…
Add support for dynamic offsets to DefaultEpilogue
inactive-30d
inactive-90d
#1274
opened Dec 19, 2023 by
ezhulenev
Loading…
Previous Next
ProTip!
Type g p on any issue or pull request to go back to the pull request listing page.