somnial

Karma: 5
Created: 8 months ago

Recent Submissions

1. ▲ Width vs. Depth: Speculating on the Margin (blog.doubleword.ai) 17 points · 11 days ago · 1 comment
2. ▲ Pushing memory bound CUDA kernels past the speed of light with data compression (fergusfinn.com) 2 points · 1 month ago · 0 comments
3. ▲ Speculative KV coding: ~4× losslessly compressed KV cache using a small model (fergusfinn.com) 2 points · 2 months ago · 0 comments
4. ▲ 70x faster cold(ish) starts for SGLang (fergusfinn.com) 1 point · 2 months ago · 0 comments
5. ▲ LLM powered data structures: A lock-free binary search tree (fergusfinn.com) 1 point · 6 months ago · 0 comments
6. ▲ Parallel Primitives for Multi-Agent Workflows (fergusfinn.com) 1 point · 6 months ago · 0 comments
7. ▲ Scheduling in LLM Inference (fergusfinn.com) 1 point · 8 months ago · 0 comments
8. ▲ How fast can an LLM go? (fergusfinn.com) 2 points · 8 months ago · 0 comments

All submissions on HN · View profile on HN