Hackernews posts about Gemma3

1. Selectively reducing eval awareness and murder in Gemma 3 27B via steering (www.lesswrong.com)
   3 points by gmays 26 days ago | discuss
2. Show HN: SycoFact 4B: Open model detecting sycophancy and delusion confirmation (huggingface.co)
   2 points by iwalton3 8 days ago | discuss
3. Show HN: Tested 12 LLMs with few-shot examples
   2 points by shuntaro-okuma 12 days ago | discuss
4. Show HN: Skillcop: Block malicious Claude Skills before they execute (github.com)
   2 points by bennydog224 18 days ago | discuss
5. Google releases Gemma 4 open models (deepmind.google)
   1805 points by jeffmcjunkin 5 days ago | 473 comments
6. Gemma 4 on iPhone (apps.apple.com)
   848 points by janandonly 1 day ago | 232 comments
7. Running Gemma 4 locally with LM Studio's new headless CLI and Claude Code (ai.georgeliu.com)
   398 points by vbtechguy 1 day ago | 98 comments
8. April 2026 TLDR Setup for Ollama and Gemma 4 26B on a Mac mini (gist.github.com)
   322 points by greenstevester 4 days ago | 119 comments
9. Show HN: Real-time AI (audio/video in, voice out) on an M3 Pro with Gemma E2B (github.com)
   271 points by karimf 1 day ago | 35 comments
10. Show HN: Gemma Gem – AI model embedded in a browser – no API keys, no cloud (github.com)
    151 points by ikessler 1 day ago | 21 comments
11. Gemma Needs Help (www.lesswrong.com)
    38 points by pr337h4m 27 days ago | 1 comments
12. Gemma 4: Byte for byte, the most capable open models (blog.google)
    24 points by meetpateltech 5 days ago | 2 comments
13. Gemma 4 running client-side in WebGPU (huggingface.co)
    8 points by astlouis44 3 days ago | discuss
14. Gemma 4 Uncensored (autoresearch results) (huggingface.co)
    5 points by adefa 1 day ago | 4 comments
15. Gemma 4 Models (huggingface.co)
    5 points by maciejgryka 5 days ago | 1 comments
16. A Visual Guide to Gemma 4 (newsletter.maartengrootendorst.com)
    4 points by mariuz 4 days ago | 1 comments
17. Run Gemma 4 in Your Browser (huggingface.co)
    4 points by oldfuture 1 day ago | discuss
18. Gemma 4 makes local AI agents practical (firethering.com)
    4 points by steveharing1 4 days ago | discuss
19. Gemma 4 running on NVIDIA and AMD from Day 0 with MAX (www.modular.com)
    4 points by sparklychipmunk 5 days ago | discuss
20. Fine-tuning Gemma 4 locally
    3 points by logicallee about 23 hours ago | discuss
21. Local LLM for OpenCode Gemma 4 26B A4B. No GPU required (grigio.org)
    3 points by grigio 3 days ago | discuss
22. Running Qwen 35B on 16GB Mac Mini via MMAP (17 tok/s) then swapping to Gemma 4 (thoughts.jock.pl)
    3 points by joozio 4 days ago | discuss
23. Gemma 4 reports, configs, and code, here are the takeaways (twitter.com)
    3 points by tzury 5 days ago | discuss
24. Gemma 4 running locally in the browser with transformers.js (huggingface.co)
    3 points by clmnt 5 days ago | discuss
25. Gemma 4 Model Card (ai.google.dev)
    3 points by Topfi 5 days ago | discuss
26. Gemma 4: The new standard for local agentic intelligence on Android (android-developers.googleblog.com)
    2 points by xnx 5 days ago | 1 comments
27. Google AI Edge Eloquent is a Gemma-powered dictation app that works offline (www.neowin.net)
    2 points by bundie about 4 hours ago | discuss
28. Gemma 4 clocking 12 tokens / SEC on Pixel 7A (twitter.com)
    2 points by amrrs about 12 hours ago | discuss
29. Welcome Gemma 4: Frontier multimodal intelligence on device (huggingface.co)
    2 points by ibobev about 19 hours ago | discuss
30. Google's Gemma 4 is good, but you won't use it (www.riteshkhanna.com)
    2 points by treadon 4 days ago | discuss