DEV Community

#cuda

Posts

đź‘‹ Sign in for the ability to sort posts by relevant, latest, or top.
Whisper turbo vs. medium on Arm and CUDA: The Faster Model Still Flipped Between CPU and GPU

Whisper turbo vs. medium on Arm and CUDA: The Faster Model Still Flipped Between CPU and GPU

Comments
11 min read
Jensen Huang quitte la direction de Nvidia : chronologie, ce que l'on sait et ce qui change (novembre 2026)

Jensen Huang quitte la direction de Nvidia : chronologie, ce que l'on sait et ce qui change (novembre 2026)

Comments
18 min read
Local LLM generation on a consumer GPU: AirColibri streams Qwen3.5-9B into modest VRAM

Local LLM generation on a consumer GPU: AirColibri streams Qwen3.5-9B into modest VRAM

Comments 2
5 min read
Rewriting My Triton FlashAttention Kernel in CuTe DSL, One Optimisation at a Time

Rewriting My Triton FlashAttention Kernel in CuTe DSL, One Optimisation at a Time

Comments
52 min read
How do you fix CUDA out of memory errors in PyTorch?

How do you fix CUDA out of memory errors in PyTorch?

Comments
10 min read
Your CUDA Code Didn't Change. Your Risk Register Just Did: What Nvidia's CEO Exit Means for Builders

Your CUDA Code Didn't Change. Your Risk Register Just Did: What Nvidia's CEO Exit Means for Builders

Comments
4 min read
The 8GB VRAM Breakthrough: How PagedAttention and Dynamic KV-Cache Quantization Unlock 64k Context on Consumer GPUs

The 8GB VRAM Breakthrough: How PagedAttention and Dynamic KV-Cache Quantization Unlock 64k Context on Consumer GPUs

Comments
11 min read
One Spark, four streams: TensorFold’s native CUDA experiment

One Spark, four streams: TensorFold’s native CUDA experiment

Comments
3 min read
Quantize or Offload? A 16GB GPU Local LLM Strategy

Quantize or Offload? A 16GB GPU Local LLM Strategy

Comments
4 min read
I built a free Windows (UEFI) & Linux GUI to unlock RTX 40/30/20 Tensor Cores (2.01x FP16/BF16/FP8 on my RTX 4070 Ti)

I built a free Windows (UEFI) & Linux GUI to unlock RTX 40/30/20 Tensor Cores (2.01x FP16/BF16/FP8 on my RTX 4070 Ti)

1
Comments
3 min read
Same A40, same Spark file: inside cortiq’s new CUDA path

Same A40, same Spark file: inside cortiq’s new CUDA path

Comments
4 min read
Splitting Network Paths with Two NICs to Get CUDA Working on RTX 5060 under WSL2 on Windows 11

Splitting Network Paths with Two NICs to Get CUDA Working on RTX 5060 under WSL2 on Windows 11

Comments
4 min read
I had an AI agent port Karpathy's GPT-2 trainer from CUDA to SYCL, and made it prove every step

I had an AI agent port Karpathy's GPT-2 trainer from CUDA to SYCL, and made it prove every step

Comments
4 min read
How Does Rust Code Run on a GPU? Tracing cuda-oxide and PTX

How Does Rust Code Run on a GPU? Tracing cuda-oxide and PTX

Comments
10 min read
How I Built an LLM Inference Engine From Scratch in CUDA (235 tokens/s on a 6 GB Laptop GPU)

How I Built an LLM Inference Engine From Scratch in CUDA (235 tokens/s on a 6 GB Laptop GPU)

Comments 1
8 min read
đź‘‹ Sign in for the ability to sort posts by relevant, latest, or top.