Hackernews posts about ModernBERT

1. 01
   

   newA Replacement for Bert: Introducing ModernBERTwww.answer.ai

   ▲ 16bitquant7ddiscusssavesaved
2. 02
   

   newShow HN: VisionLaya: Jev with Vision capabilitieshuggingface.co

   ▲ 5someguy10101016d💬 2savesaved
3. 03
   

   newShow HN: A local alternative to Jev – 94% on Banking77gist.github.com

   ▲ 3nico8ddiscusssavesaved
4. 04
   

   newShow HN: 188M Hindi encoder, 28B tokens, 8K context, 1× RTX 4090github.com

   ▲ 2kkkamur29ddiscusssavesaved
5. 05
   

   newShow HN: Context-aware Japanese furigana using Sudachi and ModernBERTwww.ezfurigana.com

   ▲ 37epitrochoid413129d💬 20savesaved
6. 06
   

   newFine-tune classifier with ModernBERT in 2025www.philschmid.de

   ▲ 19mcyc648d💬 3savesaved
7. 07
   

   newDistilling ModernBERT into a static model doesn't workminishlab.github.io

   ▲ 5stephantul614d💬 3savesaved
8. 08
   

   newShow HN: ModernBERT in Pure Cgithub.com

   ▲ 3HardikVala352d💬 1savesaved
9. 09
   

   newModernBERTarxiv.org

   ▲ 3fzliu654d💬 1savesaved
10. 10
    

    newModernBERT in Jax/Flaxgithub.com

    ▲ 3alien0006235ddiscusssavesaved
11. 11
    

    newA Replacement for Bert: Introducing ModernBERTsimonwillison.net

    ▲ 3tosh650ddiscusssavesaved
12. 12
    

    newModernBERT: Encoder-only Transformer Model Strictly Improving on past workhuggingface.co

    ▲ 2distortedsignal642ddiscusssavesaved
13. 13
    

    newA Replacement for Bert: Introducing ModernBERTwww.answer.ai

    ▲ 2CharlesW650ddiscusssavesaved
14. 14
    

    newModernBERTwww.answer.ai

    ▲ 1tosh42d💬 1savesaved
15. 15
    

    newThe release of the first ModernBERT-based embedding modelstwitter.com

    ▲ 1zanussbaum643d💬 1savesaved
16. 16
    

    newModernBERThuggingface.co

    ▲ 1tosh654d💬 1savesaved
17. 17
    

    newFine-Tune ModernBERT for RAG with Synthetic Datahuggingface.co

    ▲ 1Dowwie607ddiscusssavesaved
18. 18
    

    newTraining mRNA Language Models Across 25 Species for $165

    ▲ 148maziyar186d💬 42savesaved
19. 19
    

    newShow HN: Chonky – a neural text semantic chunking goes multilingualhuggingface.co

    ▲ 43hessdalenlight345d💬 6savesaved
20. 20
    

    newShow HN: Pi Co-pilot – Evaluation of AI apps made easywithpi.ai

    ▲ 34achintms501d💬 8savesaved
21. 21
    

    newJavelinGuard: Low-Cost Transformer Architectures for LLM Securityarxiv.org

    ▲ 29sharathr481d💬 2savesaved
22. 22
    

    newShow HN: Pi Labs – AI scoring and optimization tools for software engineersbuild.withpi.ai

    ▲ 25achintms570d💬 8savesaved
23. 23
    

    newShow HN: LettuceDetect – Lightweight hallucination detector for RAG pipelinesgithub.com

    ▲ 10justacoolname534d💬 1savesaved
24. 24
    

    newShow HN: VerbatimRAG – RAG that returns only exact text from documents

    ▲ 2justacoolname419ddiscusssavesaved
25. 25
    

    newA Replacement for BERThuggingface.co

    ▲ 348cubie654d💬 75savesaved
26. 26
    

    newShow HN: A 150M model that extracts verbatim evidence spans for RAG, no LLM callhuggingface.co

    ▲ 6justacoolname116ddiscusssavesaved
27. 27
    

    newTIL: Masked Language Models Are Surprisingly Capable Zero-Shot Learnerswww.answer.ai

    ▲ 4yurivish601ddiscusssavesaved

· edit