DeepSelect
DeepSelect: TopK kernels for DeepSeek Sparse Attention (DSA) and Samplers
Get picks like this daily. Today's launch numbers, the top launches, AI and tech news, and a daily opportunity pick, straight to your inbox.
Want to reach these readers? Sponsor an issue.
GITHUB · launched Sep 9, 2026 · 327 upvotes (top 2%) · 1 view
Niche: machine learning developer tools · 303 launches, 11 in the last 7 days ▼ -45% · Explore this niche →
Category and tags
- Horizontal
- Model & infra
- Developer
- AI-native
- Hobby / open-source project
Visit website ↗ Source listing ↗
More like this: all 303 launches in machine learning developer tools →
Could you build this?
No Developing custom TopK GPU kernels for Sparse Attention architectures requires deep low-level CUDA, Triton, and hardware-specific GPU memory hierarchy optimization.
What it would actually take: The project requires writing optimized CUDA, CUTLASS, or Triton kernels tailored to modern GPU architectures (e.g., Hopper/Blackwell) for TopK selection and sparse attention computation. It demands expertise in warp-level primitives, shared memory bank conflicts, tensor cores, and PyTorch C++ extension bindings to match peak hardware FLOPs and memory bandwidth.
Competitors
Other products that read as similar to this one : 2223 launches clear the similarity bar, closest 8 shown.
Attention rank: #59 of 2224 (itself plus its competitors, highest first, normalized so YC and Product Hunt are compared fairly).
Launched 314 days after the earliest competitor.
- DeepInfer · github · 2026-10-16 · 10428 upvotes · similarity 0.69
- KronLSTM · github · 2026-10-19 · 49 upvotes · similarity 0.65
- Simple-Attention-Sparsification · github · 2026-09-10 · 48 upvotes · similarity 0.65
- fp4Attn · github · 2026-10-25 · 15 upvotes · similarity 0.64
- FlashMask · github · 2026-10-14 · 35 upvotes · similarity 0.64
- Ascend-H3 · github · 2026-10-30 · 29 upvotes · similarity 0.63
- DASA · github · 2026-10-13 · 29 upvotes · similarity 0.63
- LeapQuant · github · 2026-11-11 · 50 upvotes · similarity 0.62
The niche, labels, vibe code rating and competitor comparison on this page are analysis by Nicheloom. If you use them, please credit nicheloom.com. How you may use our data.