Oops! That page is private.

Take me home

Popular topics

I am EXTREMely disappointed with the current state of DGX SparkDGX Spark / GB10
Qwen3.5-122B-A10B on single Spark: up to 51 tok/s (v2.1 — patches + quick-start + benchmark)DGX Spark / GB10
Why Turboquant saves DGX twiceDGX Spark / GB10
Should we as a community gofundme one Spark for Eugr’s nightly builds?DGX Spark / GB10
NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4DGX Spark / GB10
Gemma 4 Day-1 Inference on NVIDIA DGX Spark — Preliminary BenchmarksDGX Spark / GB10
Small monitor program for DGX SparkDGX Spark / GB10 Projects
NemoClaw on SparkDGX Spark / GB10
50%+ Improvement on spark?!DGX Spark / GB10
Unsloth Studio, GUI to train modelsDGX Spark / GB10
More…

Recent topics

Simple NVTX annotation breaks profiling report (Missing CUDA HW row)Profiling Linux Targets
cuptiProfilerGetCounterAvailability Causes a Segmentation Fault with Cuda Toolkit 13.0.0 When Using Dynamic Shared LibrariesCUPTI – CUDA Profiler Tools Interface
NVFORTRAN ignores options to compile without AVX2 instructions (SandyBridge processor)nvc, nvc++ and nvfortran
Anyone have a solution for LoRA training of recent MoE models like Qwen3.5-35B-A3B or Gemma-4-26B-A4B *and* successfully running in vLLM?DGX Spark / GB10
Gemma 4 VLM VRAM/Host Memory Leak — Full Investigation ReportCUDA Programming and Performance
TensorRT support for RTX 5090TensorRT for RTX
Built a Peta-scale out-of-core PyTorch engine on an 8GB laptop GPU that processes a 150GB dataset into 130GB of geometry using inverted batch-streaminCUDA Programming and Performance
cuBLAS batched FP32 SGEMM dispatcher picks suboptimal kernel on RTX 5090 (sm_120)CUDA Programming and Performance
My GUI is gone and Nvidia smi is not workingDGX Spark / GB10
NIM Access Entitlement CertificateDGX Spark / GB10
More…