HN Explorer

BUFU

↗ HN profile [ 47.8 ms ] full profile
Karma
0
Created
()
Submissions
0
  1. Qualcomm acquires Nexa AI, open-sources GenAI runtime for Hexagon NPUs (github.com)
    1 points 1
  2. We Ran GPT‑OSS 20B Local on a Phone (nexa.ai)
    1 points 0
  3. Qwen3-VL-30B-A3B-Instruct and Thinking (huggingface.co)
    6 points 0
  4. New Engine to Run SOTA AI Models on Qualcomm NPU Across Phone, PC, Cars, and IoT (nexa.ai)
    1 points 0
  5. First vision language model built off Open AI GPT-OSS (huggingface.co)
    3 points 0
  6. First Multimodal AI Model Designed for NPUs (huggingface.co)
    1 points 0
  7. Nexa AI Blogs (nexa.ai)
    1 points 0
  8. LFM2-VL: Efficient Vision-Language Models (liquid.ai)
    3 points 0
  9. Stanford CS336 Language Modeling from Scratch (youtube.com)
    19 points 0
  10. Ollama's new app (ollama.com)
    560 points 69
  11. You can now connect a directory of apps and tools to Claude with one click (claude.ai)
    2 points 1
  12. Ask HN: What tools have you tried to run AI locally on mobile?

    Everyone has a phone, and it is the place where we need most privacy. Who have tried running LLMs on mobile or built local AI projects on mobile? Out of curiosity: 1. What tools have you tried? 2. What specific step…

    2 points 0
  13. A C++ library to efficiently run Gemma-3N across various platform (github.com)
    5 points 0
  14. The Trump-Musk feud has been great for X, which jumped up the App Store charts (techcrunch.com)
    6 points 2
  15. How we’re responding to The NYT’s data demands in order to protect user privacy (openai.com)
    284 points 343
  16. ChatGPT Deep Research connects cloud apps (twitter.com)
    1 points 0
  17. 1 points 0
  18. Local AI generates highly realistic dialogue from a transcript (yummy-fir-7a4.notion.site)
    3 points 2
  19. Shroud's Spectre Divide and its developer are shutting down (theverge.com)
    1 points 0
  20. What Went Wrong with Skype? (theverge.com)
    4 points 3
  21. Anthropic's Recommendations to OSTP for the U.S. AI Action Plan (anthropic.com)
    3 points 1
  22. 1 points 1
  23. Quantized DeepSeek R1 Distill Models with Original Model Accuracy (nexa.ai)
    2 points 0
  24. Multimodal Model Quantization Support Through LLM Compressor by Neural Magic (neuralmagic.com)
    1 points 0
  25. DeepSeek-R1-Distill-Qwen-1.5B Surpasses GPT-4o in certain benchmarks (huggingface.co)
    39 points 17