Search results

Clarification regarding 3D Transposed Convolution speedup (Section 5.3)

Status: Open.

#18 In deepreinforce-ai/CUDA-L1;

· Dasor opened on Apr 23, 2026

Request Pytorch and NVCC Setup

Status: Open.

#17 In deepreinforce-ai/CUDA-L1;

· LemonAndRabbit opened on Dec 14, 2025

Are the released H100 kernels trained on H100 or a100 ?

Status: Open.

#16 In deepreinforce-ai/CUDA-L1;

· zacharyb3 opened on Nov 25, 2025

Is there any way to gurranttee kernel correctness rather than only relying on torch.allclose?

Status: Open.

#14 In deepreinforce-ai/CUDA-L1;

· GeorgeTay92 opened on Nov 18, 2025

Why is it referred to as L1? 🤔 Is there a deeper significance behind the name?

Status: Open.

#13 In deepreinforce-ai/CUDA-L1;

· SaladDay opened on Nov 3, 2025

🐛 Issue: Lazy Evaluation Bypass in eval_cuda.py Performance Measurement

Status: Open.

#12 In deepreinforce-ai/CUDA-L1;

· BUAA-CLab opened on Oct 27, 2025

cuda-L1 for V100

Status: Open.

#10 In deepreinforce-ai/CUDA-L1;

· valentijnvenus opened on Aug 5, 2025

cuda-L1 for 5090/4090

Status: Open.

#9 In deepreinforce-ai/CUDA-L1;

· anujsinha72094 opened on Aug 4, 2025

Release CUDA-L1 model on Hugging Face

Status: Open.

#8 In deepreinforce-ai/CUDA-L1;

· NielsRogge opened on Jul 31, 2025

Full end to end pipeline open source?

Status: Open.

#6 In deepreinforce-ai/CUDA-L1;

· aymuos15 opened on Jul 22, 2025