Hackernews posts about FP8 GEMM

1. Show HN: Doom running in OpenSCAD at 10-20 FPS (www.mikeayles.com)
   2 points by mikeayles 28 days ago | 1 comments
2. Tell HN: OpenAI's Codex CLI is currently free to use
   6 points by davidpolberger 18 days ago | 1 comments
3. DeepGEMM: clean and efficient FP8 GEMM kernels with fine-grained scaling (github.com)
   391 points by mfiguiere 12 months ago | 67 comments
4. DeepSeek Open Source DeepGEMM – FP8 GEMM Library(300 lines for 1350+ FP8 TFLOPS) (twitter.com)
   4 points by helloericsf 12 months ago | 1 comments
5. Show HN: Fine-tuned Llama 3.2 3B to match 70B models for local transcripts (bilawal.net)
   31 points by phantompeace 6 months ago | 8 comments
6. Transformers 2.0: What Ilya and Sam Might Have Missed
   5 points by metehan777 over 1 year ago | 1 comments
7. Explosive GEMM: arbitrarily large FP error can be incurred in the GEMM operation (github.com)
   1 points by statusfailed 2 months ago | discuss
8. Ozaki Scheme II: A GEMM-oriented emulation of FP matrix multiplication using INT (arxiv.org)
   2 points by matt_d 10 months ago | discuss
9. Show HN: KiDoom – Running DOOM on PCB Traces (www.mikeayles.com)
   362 points by mikeayles 3 months ago | 49 comments
10. Show HN: Realtime, expressive AI personas that you can video call (playground.keyframelabs.com)
    4 points by kraddypatties 3 months ago | 2 comments
11. Show HN: I built an MCP to help build Cursor for Video Editing (smithery.ai)
    2 points by misbahsy 9 months ago | discuss
12. Show HN: VerityNgn–Open-source AI that fact-checks YouTube videos (hotchilianalyticsllc.mintlify.app)
    1 points by ajjcoppola about 1 month ago | discuss