ModernBERT
Fewer than 5 people recommend it so far, so there's no written verdict. Here is what commenters said.
3recommend
0warn against
Mar 3, 2026first mention
Oct 11, 2026latest mention
Why people recommend it
“ModernBERT is amazing if you haven’t used it before, worth spending time with”
“ModernBERT was the clear winner. It learned faster and performed the task better by a significant margin”
Show 2 more
“embeddings are easy to fine tune. Try modern bert.”
“small encoder-only classifiers (DeBERTa-small, ModernBERT)”
Caveats and dissent
No one in these comments warned against it.
Who recommends it in their own words
Threads where it came up
- Build your own decision model 1 mention
- Jeff – Jev-compatible 0.8B decision models, trained at home, ~30 ms 1 mention
- Show HN: Hacker News, without AI 1 mention
- Apple Neural Engine: Architecture, Programming, and Performance 1 mention
- Good results fine tuning a local LLM like Qwen 3:0.6B to categorize questions 1 mention
- Ask HN: Best Embedding Models? 1 mention
Data: modernbert.json. Quotes are verbatim; the verdict and theme labels are written by a model from these comments, and every clause links to its sources.
All comments 7
The real guide is here: https://huggingface.co/convaiinnovations/laya
“using the same ModernBERT stuff”
“a Modern Bert-based classifier fine-tuned to detect AI-related content”
“ModernBERT is amazing if you haven’t used it before, worth spending time with”
“ModernBERT was the clear winner. It learned faster and performed the task better by a significant margin”
“embeddings are easy to fine tune. Try modern bert.”
“small encoder-only classifiers (DeBERTa-small, ModernBERT)”