Hackernews posts about Gemma 3 270

  1. I've tested some local LLMs on prosumer hardware, here are some findings
    7 points by felineflock 28 days ago | discuss
  2. Show HN: Don't trust me bro: fixing GPT-OSS (3.49B tokens, 1k GPU hours, 1x3090) (github.com)
    2 points by iamskeole 19 days ago | discuss
  3. Gemma 3 270M: Compact model for hyper-efficient AI (developers.googleblog.com)
    824 points by meetpateltech about 1 year ago | 317 comments
  4. Gemma 3 270M re-implemented in pure PyTorch for local tinkering (github.com)
    417 points by ModelForge about 1 year ago | 57 comments
  5. Own your AI: Learn how to fine-tune Gemma 3 270M and run it on-device (developers.googleblog.com)
    5 points by simonpure 11 months ago | discuss
  6. Gemma 3-270M (huggingface.co)
    4 points by georgehill about 1 year ago | discuss
  7. You can fine-tune Gemma3-270M and prepare for secure deployment within minutes (www.youtube.com)
    3 points by astro_09 about 1 year ago | discuss
  8. Gemma 3 270M – Google's NEW AI – How to Fin-tune Gemma3 [video] (www.youtube.com)
    2 points by rjn32s about 1 year ago | 1 comments
  9. Tiny gemma3 270M implementation in R (github.com)
    2 points by dfalbel about 1 year ago | discuss
  10. Gemma 3 270M (twitter.com)
    2 points by tosh about 1 year ago | discuss
  11. Launch HN: Cactus (YC S25) – AI inference on smartphones (github.com)
    123 points by HenryNdubuaku about 1 year ago | 63 comments
  12. Show HN: AfriTales – Discover the Magic of African Storytelling (afritales.org)
    42 points by ggap about 1 year ago | 26 comments
  13. Ask HN: Any Local LLM can I run without GPU for Local Agentic workflow AI?
    8 points by limondas 3 months ago | 4 comments
  14. Show HN: Nabu (TTS Reader and LLM Playground on Android) (github.com)
    7 points by mewmix about 1 year ago | discuss
  15. Running a 270M LLM on Android (architecture and benchmarks)
    2 points by ayushranjan99 10 months ago | discuss
  16. Show HN: Go LLM inference with a Vulkan GPU back end that beats Ollama's CUDA (github.com)
    1 points by computerex 7 months ago | discuss
  17. Gemma 3 QAT Models (simonwillison.net)
    4 points by mdp2021 over 1 year ago | 3 comments
  18. Google's Gemma 3 is an open source, multimodal single-GPU AI 128K context window (arstechnica.com)
    4 points by danboarder over 1 year ago | 1 comments
  19. Show HN: Interactive semantic map and analysis of Hacker News stories in 2025 (lincolnmaxwell.com)
    2 points by youngermax 8 months ago | discuss
  20. Fine-tuning Gemma 3 for mobile (opensource.googleblog.com)
    1 points by nkko 9 months ago | discuss
  21. Show HN: Local LLM Notepad – run a GPT-style model from a USB stick (github.com)
    40 points by davidye324 about 1 year ago | 9 comments
  22. Show HN: Fine-tuned Llama 3.2 3B to match 70B models for local transcripts (bilawal.net)
    31 points by phantompeace about 1 year ago | 8 comments
  23. Show HN: Tested 12 LLMs with few-shot examples
    2 points by shuntaro-okuma 6 months ago | discuss
  24. Ask HN: How do you use Local LLMs? (April 2026)
    3 points by Hixon10 5 months ago | discuss
  25. Tq-KV – Rust implementation of TurboQuant that works on GGUF models
    3 points by onurgokyildiz 6 months ago | discuss
  26. Show HN: Cactus Hybrid: We taught Gemma 4 to know when it's wrong (github.com)
    191 points by HenryNdubuaku about 2 months ago | 44 comments
  27. Run Gemma 4 locally with OpenClaw in 3 steps (twitter.com)
    2 points by xnx 6 months ago | discuss
  28. Show HN: Glq LLM quantization using E8 lattice (github.com)
    2 points by acd 4 months ago | discuss
  29. Show HN: Gemma 4 26B A4B running on an iPhone 17 Pro via model paging (noemaai.com)
    3 points by armin976 about 2 months ago | 2 comments
  30. New Inference Server for DGX Spark: large model C4:55-90 tok/s no spec decode
    4 points by medicis123 about 2 months ago | discuss
  31. Presentation Decks are a lot easier to create with Gamma
    40 points by FutureStacked_ about 1 year ago | 2 comments
  32. Gamma-ray pulsations reveal extreme 3.2-ms pulsar 3,900 light-years away (phys.org)
    4 points by wglb about 2 months ago | 2 comments