Hackernews posts about bfloat16

1. llm.c: multi-GPU, bfloat16, flash attention, ~7% faster than PyTorch (twitter.com)
   121 points by tosh about 1 year ago | 10 comments
2. Bfloat16 support coming to Apple's Metal and PyTorch [video] (developer.apple.com)
   93 points by dlewis1788 almost 2 years ago | 53 comments
3. Supporting BFLOAT16 in Fortran: "Not Recommended"? (fortran-lang.discourse.group)
   7 points by ggraphilia over 1 year ago | 2 comments
4. Evaluation of Bfloat16, Posit, and Takum Arithmetics in Sparse Linear Solvers (arxiv.org)
   3 points by jobstijl 5 months ago | 2 comments
5. BFloat16: The secret to high performance on Cloud TPUs (cloud.google.com)
   3 points by mfiguiere over 1 year ago | 1 comments
6. RISC-V Support for BFloat16 (fprox.substack.com)
   3 points by camel-cdr almost 2 years ago | discuss
7. RISC-V with Linux 6.15 Adds Support for BFloat16 "BF16" Instructions (www.phoronix.com)
   2 points by jrepinc 2 months ago | discuss
8. llama.cpp bfloat16 support (github.com)
   2 points by indigodaddy about 1 year ago | discuss
9. Supporting BFLOAT16 in ISO_Fortran_env: not recommended (j3-fortran.org)
   1 points by zaikunzhang over 1 year ago | 7 comments
10. Converging Harmonic Series with bfloat16 (github.com)
    1 points by octocop over 1 year ago | 1 comments
11. Scale 1.3 Adds BFloat16 and New Features for Compiling CUDA Apps on AMD GPUs (www.phoronix.com)
    1 points by latchkey about 2 months ago | discuss
12. Show HN: Dive into Transformers and LLM World – Llama 3.1 in Go, Step by Step (github.com)
    4 points by adalkiran 10 months ago | 2 comments
13. Show HN: LLaMA Nuts and Bolts, A holistic way of understanding how LLMs run (github.com)
    3 points by adalkiran about 1 year ago | discuss
14. Show HN: Retrieval Augmented Generation Optimised LLM's (huggingface.co)
    2 points by alextttty almost 2 years ago | discuss
15. Ask HN: Clarify VRAM usage during LLM forward pass
    1 points by furiousteabag over 1 year ago | discuss
16. WebGPU in Chrome Canary now supports float16 for faster local LLM inference (twitter.com)
    18 points by NiekvdMaas almost 2 years ago | 5 comments
17. TC39 advances proposals for RegExp Escaping, Float16Array, Redeclarable vars (socket.dev)
    7 points by feross 4 months ago | discuss
18. Float16Array Turbofan Pipeline DevLog – Weekly Sync with V8 Leader (blog.seokho.dev)
    1 points by devsdk 4 months ago | discuss
19. Launch HN: Deepsilicon (YC S24) – Software and hardware for ternary transformers
    189 points by areddyyt 9 months ago | 79 comments
20. Show HN: Model2Vec: make sentence transformers 500x faster on CPU, 15x smaller (github.com)
    9 points by stephantul 9 months ago | 2 comments
21. Show HN: Model2Vec: make sentence transformers 500x faster on CPU, 15x smaller (github.com)
    6 points by stephantul 9 months ago | 2 comments
22. Tell HN: I learnt some new things while working on my project
    1 points by behnamoh almost 2 years ago | discuss
23. Show HN: LLMs API for Asian languages. 90% cheaper than GPT-3.5 (float16.cloud)
    2 points by mattick27 over 1 year ago | discuss
24. Show HN: LLM Service for Asian Languages (float16.cloud)
    1 points by mattick27 over 1 year ago | discuss