Hackernews posts about MLX

  1. Apples to Apples: MLX vs. Llama.cpp for Gemma 4 12B on an M1 16GB (ziraph.com)
    6 points by ABS 16 days ago | 1 comments
  2. WWDC26: Run local agentic AI on the Mac using MLX [video] (www.youtube.com)
    5 points by jorisw 11 days ago | 1 comments
  3. Run local agentic AI on the Mac using MLX [video] (www.youtube.com)
    4 points by angristan 12 days ago | discuss
  4. GGUF vs. MLX when to choose what when running Local LLMs (muhammadraza.me)
    4 points by mr_o47 17 days ago | discuss
  5. Mlx-optiq: per-layer mixed-precision LLM quantization for Apple Silicon (mlx-optiq.com)
    2 points by codelion 7 days ago | discuss
  6. Run local agentic AI on the Mac using MLX (WWDC 2026) [video] (developer.apple.com)
    2 points by sebiw 9 days ago | discuss
  7. Improving LM Studio's MLX Engine for Agentic Workflows (twitter.com)
    2 points by tosh 15 days ago | discuss
  8. Show HN: Ported Cerebras REAP to MLX – Prune MoE Experts on a MacBook (github.com)
    2 points by egesabanci 20 days ago | discuss
  9. Benchmarking TurboQuant with MLX on Apple Silicon (www.youtube.com)
    2 points by tcp_handshaker 21 days ago | discuss
  10. Mlx-Code: A Coding Agent That Speaks Git Natively (www.mlx-code.com)
    1 points by JosefAlbers 4 days ago | discuss
  11. Show HN: iPhone ANE holds LLM tok/s while MLX and LiteRT thermal-throttle (github.com)
    1 points by mlboy 18 days ago | discuss
  12. Show HN: Omni – Local-first multimodal file search on macOS (hanxiao.io)
    6 points by artex_xh 16 days ago | 2 comments
  13. Ask HN: What are some good/fast coding models for Apple Silicon?
    2 points by LoganDark 5 days ago | 3 comments
  14. A 1B humanizer that matches human writing on an AI detector (mlx-optiq.com)
    3 points by codelion 20 days ago | discuss
  15. I indexed 669 GB of my GoPro videos using my M1 Max computer and local ML models
    438 points by iliashad 7 days ago | 115 comments
  16. Chipotlai Max (github.com)
    401 points by nigelgutzmann 20 days ago | 66 comments
  17. Maxproof (arxiv.org)
    137 points by ilreb 9 days ago | 13 comments
  18. Qwen3.7-Max Ran for 35 Hours on Unknown Hardware and Achieved a 10× Speedup (firethering.com)
    38 points by steveharing1 27 days ago | 31 comments
  19. Darknet Market Maximalism (antimoonboy.com)
    32 points by Cider9986 18 days ago | discuss
  20. Claude Opus 4.8 Max responding to an empty message (xcancel.com)
    27 points by frb 18 days ago | 4 comments
  21. Everything I Learned Training Frontier Small Models – Maxime Labonne, Liquid AI [video] (www.youtube.com)
    16 points by Topfi 4 days ago | discuss
  22. LFM2.5-8B-A1B: An Even Better On-Device Mixture-of-Experts (www.liquid.ai)
    11 points by pember 24 days ago | 3 comments
  23. Tell HN: Max messenger app removed from App Store
    9 points by secondary_op 18 days ago | 4 comments
  24. Reconstructing a Mixbook movie from its data API with FFmpeg (segar.me)
    9 points by msegar 29 days ago | 2 comments
  25. Anthropic_API_key? Anthropic will bill your API account instead of your Max plan (old.reddit.com)
    7 points by behnamoh 15 days ago | 2 comments
  26. Jury clears Boeing in 737 MAX fraud lawsuit (www.reuters.com)
    7 points by josephwegner 29 days ago | 1 comments
  27. Anthropic Sued over Alleged False Advertising on Claude Max Subscription Limits (www.cnet.com)
    7 points by SilverElfin 6 days ago | discuss
  28. Protein name confusion created antibody mix-up affecting papers (www.science.org)
    7 points by Brajeshwar 15 days ago | discuss
  29. Hundreds of cancer papers presented incorrect data after p16 protein mixup (forbetterscience.com)
    6 points by ilamont 18 days ago | discuss
  30. Show HN: FusionHarness – An Open-source Mixture-of-Agents compound-model server (github.com)
    5 points by jackxlau 4 days ago | 1 comments
More