Tags
Tags give the ability to mark specific points in history as being important
v6-preview.62
dcc6bb8f
·
Add minimal cuSPARSE (#621)
·
Mar 13, 2026
v6-preview.61
796ad6ca
·
Support some cublaslt settings required by COEIROINK (#619)
·
Mar 02, 2026
v6-preview.60
a3b322f6
·
Add various bits and pieces required by pytorch (#615)
·
Mar 02, 2026
v6-preview.59
bd826944
·
Support cuMemAllocPitch_v2 and cuMemcpy2D_v2 (#616)
·
Feb 20, 2026
v6-preview.58
e68e8553
·
Enable handle from cublasCreate to be used in cublasLt calls (#587)
·
Feb 17, 2026
v6-preview.57
e911230e
·
Finish extended precision arithmetic (#609)
·
Feb 16, 2026
v6-preview.56
58d90b0b
·
Fix loading CUDA modules from dark api and add a tool to verify Windows library loading (#612)
·
Feb 16, 2026
v6-preview.55
a5ecf6ae
·
ctlz bug fix (#610)
·
Jan 20, 2026
v6-preview.54
0faa099c
·
Implement extended precision integer addition (#607)
·
Jan 20, 2026
v6-preview.53
59012649
·
Host functions for vLLM (#606)
·
Jan 15, 2026
v6-preview.52
8e1f9953
·
Add sad and dp2a instructions (#605)
·
Jan 15, 2026
v6-preview.51
2beaee40
·
Use partial parsing result in release mode (#603)
·
Jan 13, 2026
v6-preview.50
b3ede5d8
·
Update docs (add llama.cpp, zluda_precompile sections) (#602)
·
Jan 12, 2026
v6-preview.49
0d7f3fd2
·
Stop failing on bf16 uint_to_fp on amdgpu < gfx11 (#601)
·
Jan 12, 2026
v6-preview.48
d9a5304f
·
Add CUDA 13.1 compatibility (#599)
·
Jan 09, 2026
v6-preview.46
7fe40d35
·
When building ZLUDA in CI, make sure we build Linux binaries compatible with...
·
Jan 08, 2026
v6-preview.47
7fe40d35
·
When building ZLUDA in CI, make sure we build Linux binaries compatible with...
·
Jan 08, 2026
v6-preview.44
294f2364
·
Allow implicit conversion from bit scalar to vec for st (#585)
·
Jan 03, 2026
v6-preview.45
294f2364
·
Allow implicit conversion from bit scalar to vec for st (#585)
·
Jan 03, 2026
v6-preview.43
0254f11c
·
Implicitly convert from float to bit type (#583)
·
Jan 03, 2026
1
2
3
4
5
…
9