|
Prime k-tuplet sieve in CUDA on an RTX 5060 Ti: new record prime 18-tuplet
|
|
16
|
244
|
November 15, 2026
|
|
PRPLL NTT now supports both CUDA and OpenCL
|
|
8
|
812
|
November 14, 2026
|
|
The CUDA version of nvcc --version is 11.2, and the nvidia-smi version is 12.6
|
|
4
|
608
|
November 14, 2026
|
|
CUDA GPU not getting detected after Python environment update
|
|
1
|
96
|
November 13, 2026
|
|
Meaning of FuncExecutionModel kernel launch (CUPTI) attribute
|
|
1
|
30
|
November 12, 2026
|
|
K80, Is it possible to still use these cards?
|
|
14
|
6428
|
November 11, 2026
|
|
CUDA error: no kernel image is available for execution on the device
|
|
1
|
131
|
November 11, 2026
|
|
The CUDA 13.4 nvcc compiler still cannot enable the C++23 standard on Windows
|
|
8
|
409
|
November 11, 2026
|
|
Announcement: cuda-api-wrappers 0.9.0 - released
|
|
0
|
40
|
November 10, 2026
|
|
RTX 4060 Laptop: BF16 torch.bmm illegal memory access across CUDA 12.4 and 13.0
|
|
3
|
73
|
November 10, 2026
|
|
NVCC 13.1.x - 13.5.37 crashes in cudafe++ on a nested enum default member initializer in a template-local class
|
|
0
|
42
|
November 8, 2026
|
|
GB206 SM120 — FMLA32 speed-select mapping and supported read-only observation
|
|
7
|
162
|
November 6, 2026
|
|
Bit-reproducible MXFP4 GEMV on an RTX 4060 (sm_89): how to make it faster without changing the summation order?
|
|
10
|
206
|
November 6, 2026
|
|
Hook cudaLaunchKernel failed
|
|
1
|
36
|
November 6, 2026
|
|
CUDA cuLibraryGetUnifiedFunction missing example to show how to use those api
|
|
1
|
59
|
November 6, 2026
|
|
Diagnosing 4.6 ms CUDA RNS/NTT latency on RTX PRO 6000 Blackwell
|
|
1
|
51
|
November 5, 2026
|
|
Two H100 SXMs differ by 15% on large BF16 matmuls but not on medium sizes. Is this normal?
|
|
0
|
41
|
November 5, 2026
|
|
Running cuda-checkpoint on SGLang fails on H200, but succeeds on H100
|
|
2
|
362
|
November 2, 2026
|
|
Enable NVFP4 Training on RTX 50-Series and PRO 6000 — Boost Performance and VRAM Efficiency
|
|
0
|
166
|
November 2, 2026
|
|
Is pypi.ngc.nvidia.com Python Repository Decommissioned?
|
|
0
|
141
|
October 27, 2026
|
|
Compressed-resident genomic sequence retrieval: 558 HPRC assemblies, CPU/CUDA archive compatibility
|
|
7
|
180
|
October 26, 2026
|
|
How to stridedly read data for Tensor core?
|
|
7
|
460
|
October 23, 2026
|
|
CUDA graph hangs with cross-green-context edges on R610 driver; works on R580
|
|
1
|
97
|
October 23, 2026
|
|
Request: Try out the new cuda-api-wrappers 0.9.0 release candidate
|
|
0
|
56
|
October 23, 2026
|
|
Cannot download repodata RHEL9 repo
|
|
2
|
639
|
October 23, 2026
|
|
Bypassing the Cloud: Deploying a Heterogeneous Multi-GPU Private Inference Cluster on Bare-Metal Debian
|
|
1
|
131
|
October 23, 2026
|
|
Incorrect Checksum in CUDA 13.4.1 Documentation
|
|
0
|
35
|
October 23, 2026
|
|
How are host-to-device copies split across copy engines on Blackwell? Can we get in-order PCIe reads from peer/IO memory?
|
|
0
|
60
|
October 22, 2026
|
|
A correctly rounded implementation of the half-precision logarithmic function hlog1p()
|
|
1
|
92
|
October 22, 2026
|
|
GPU resource calculator
|
|
8
|
2724
|
October 21, 2026
|