Hackernews posts about Gemma

1. Accelerating Gemma 4: faster inference with multi-token prediction drafters (blog.google)
   687 points by amrrs 17 days ago | 330 comments
2. Indexing a year of video locally on a 2021 MacBook with Gemma4-31B (50GB swap) (blog.simbastack.com)
   448 points by asenna 1 day ago | 127 comments
3. Show HN: Local LLM code-generation with Gemma 4 e2B via JSON AST to Clojure (github.com)
   19 points by vegnus 3 days ago | discuss
4. Grafting a Speech Head onto Gemma 4 E4B (www.frisson-labs.com)
   3 points by ymaws 14 days ago | 1 comments
5. Show HN: Airplane AI – Local NDA Safe AI Powered by Gemma (airplane-ai.franzai.com)
   3 points by franze 14 days ago | discuss
6. On-Device AI Coming to React Native with Gemma and React Native Executorch (twitter.com)
   3 points by kuligkar 17 days ago | discuss
7. A weekend with LoRA on Gemma 4 E2B: instrumenting what fine-tuning changes (aiexplr.com)
   3 points by mailharishin 26 days ago | discuss
8. Gemma 4 architecture support for QVAC-Fabric (Tether's llama.cpp fork) (github.com)
   2 points by oliveskin 23 days ago | discuss
9. How to run a local coding agent with Gemma 4 and Pi (patloeber.com)
   2 points by mariuz 24 days ago | discuss
10. Running Gemma 4 31B on Mac with Ollama (sammyrulez.github.io)
    2 points by sammy_rulez 26 days ago | discuss
11. Why Gemma-4 26B MoE works in HuggingFace but breaks in prod inference engines (github.com)
    1 points by maeddesg 7 days ago | discuss
12. Gemma Chat: Offline Vibe Coding on Apple Silicon (github.com)
    1 points by steveharing1 13 days ago | discuss
13. New Gemma chat template update by Google (huggingface.co)
    1 points by whythismatters 17 days ago | discuss
14. Using Google's Gemma 4 E4B Local AI Model to Reverse Engineer a Simple Crackme (github.com)
    1 points by nekitamo 25 days ago | discuss
15. Show HN: Statewright – Visual state machines that make AI agents reliable (github.com)
    126 points by azurewraith 10 days ago | 59 comments
16. Show HN: Site Mogging (sitemogging.com)
    70 points by jilles 21 days ago | 77 comments
17. Show HN: A new benchmark for testing LLMs for deterministic outputs (interfaze.ai)
    60 points by khurdula 23 days ago | 30 comments
18. Show HN: ChonkLM – Tiny language models running offline in the browser (chonklm.com)
    6 points by bilalba 13 days ago | discuss
19. Ask HN: Open-Source Coding Model and Harness at Claude Sonnet / Opus Level Perf?
    3 points by connecteev 29 days ago | 3 comments
20. Show HN: I blind-tested 14 LLMs on a WP plugin task. Surprising Findings (github.com)
    3 points by guilamu 29 days ago | 2 comments
21. Show HN: Charm – on-device spelling, grammar, and prediction for macOS (www.theodorehq.com)
    3 points by theodorehq about 4 hours ago | 1 comments
22. Show HN: I read Replika's privacy policy and then built a competitor (apps.apple.com)
    3 points by krunkworx 26 days ago | discuss
23. Ask HN: Will local models on normal hardware ever compete?
    1 points by locusofself 26 days ago | 3 comments
24. Germany Overtakes US in Ammunition Production Capacity (www.newsweek.com)
    161 points by vrganj 23 days ago | 203 comments
25. Germany has become the largest ammunition producer in the world (prm.ua)
    140 points by doener 23 days ago | 107 comments
26. Germany's Sovereign Tech Fund Backs KDE with €1.3M (www.theregister.com)
    125 points by Lihh27 8 days ago | 18 comments
27. CODA: Rewriting Transformer Blocks as GEMM-Epilogue Programs (arxiv.org)
    96 points by matt_d about 14 hours ago | 12 comments
28. German intelligence offices snub Palantir software (www.dw.com)
    76 points by abawany 8 days ago | 29 comments
29. U.S. to Withdraw 5k Troops from Germany, Pentagon Says (www.nytimes.com)
    62 points by mikhael 21 days ago | 47 comments
30. Germany goes from labour shortages to hiring freezes (www.ft.com)
    55 points by doener 4 days ago | 129 comments