ModernBERT
Ranked 6 of 8 embedding models by recommenders
Free open encoder model that people fine-tune for classification and embeddings; it learns quickly and works well as a starting point. Critics say it underperforms on some tasks.
Free open encoder model that people fine-tune for classification and embeddings; it learns quickly and works well as a starting point. Critics say it underperforms on some tasks.
9 people gave an opinion
- recommend it
- 7
- criticize it
- 2
- approval share
- 78%
Recommended in 5 of 13 months
- November 2025: not recommended
- December 2025: not recommended
- January 2026: not recommended
- February 2026: not recommended
- March 2026: recommended
- April 2026: not recommended
- May 2026: recommended
- June 2026: recommended
- July 2026: recommended
- August 2026: not recommended
- September 2026: recommended
- October 2026: not recommended
- November 2026: not recommended (month in progress)
Why it doesn’t carry the badge
- 7 people recommend itNeeds 10 or more, not met.
- 3.5 recommenders for every criticNeeds 2 to 1 or better, met.
- Recommended in 5 of 13 monthsNeeds 4 or more, met.
The same three rules apply to every product, and the badge can’t be bought. How the badge works
What people say about it
The reasons people give, with the number of comments that give each one. Open a reason to read what they said.
Why people recommend it
Each mark is one comment
Learns quickly and beats alternatives
“ModernBERT was the clear winner. It learned faster and performed the task better by a significant margin”
Comment from 22 Jun 2026 in “Good results fine tuning a local LLM like Qwen 3:0.6B to categorize questions” “ModernBERT is amazing if you haven't used it before, worth spending time with”
Comment from 30 Jun 2026 in “Apple Neural Engine: Architecture, Programming, and Performance” Good starting point for fine-tuning
“it is very fruitful to start with an off-the-shelf embedding model like ModernBERT”
Comment from 29 Sep 2026 in “Jeff – Jev-compatible 0.8B decision models, trained at home, ~30 ms” “Fine-tuning modernbert works even better but probably won't meet the criteria”
Comment from 30 Mar 2026 in “An NSFW Filter for Marginalia Search” “embeddings are easy to fine tune”
Comment from 5 May 2026 in “Ask HN: Best Embedding Models?” Useful embeddings for classical algorithms
“ModernBERT for feature generation plus classical algorithms from scikit-learn works so well”
Comment from 10 Jul 2026 in “Ask HN: What are you running locally on your machine with LLMs?” “it is very fruitful to start with an off-the-shelf embedding model like ModernBERT”
Comment from 29 Sep 2026 in “Jeff – Jev-compatible 0.8B decision models, trained at home, ~30 ms”
What people warn about
Each mark is one comment
Underperforms on some tasks
“Anything that required knowledge / familiarity mmBERT and ModernBERT post-trains performed much worse.”
Comment from 21 Sep 2026 in “Kev: Tiny Jev-like family of decision models built on top of Qwen3.5” “ModernBERT didn't have the learning capacity for a single problem”
Comment from 21 Sep 2026 in “Kev: Tiny Jev-like family of decision models built on top of Qwen3.5”
Other embedding models people recommend
The four most recommended other embedding models, with the same counts for each, so you can compare like for like. The tick on each bar marks the 2 to 1 line the badge needs.
| Product | Recommend | Criticize | Approval | Months | Owned | Badge |
|---|---|---|---|---|---|---|
| EmbeddingGemma | 27 | 4 | 87% | 3 | — | — |
| Ternlight | 17 | 3 | 85% | 1 | — | — |
| CLIP by OpenAI | 10 | 5 | 67% | 6 | — | Approved |
| Gemini Embedding 2 | 9 | 3 | 75% | 3 | — | — |
| ModernBERTThis page | 7 | 2 | 78% | 5 | — | — |
What people said
9 of the 10 comments that mention it, newest first. Each links to the original comment.
“it is very fruitful to start with an off-the-shelf embedding model like ModernBERT”
“ModernBERT didn't have the learning capacity for a single problem”
“Anything that required knowledge / familiarity mmBERT and ModernBERT post-trains performed much worse.”
“ModernBERT for feature generation plus classical algorithms from scikit-learn works so well”
“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”
“would probably map query to category automatically”
“embeddings are easy to fine tune”
“Fine-tuning modernbert works even better but probably won't meet the criteria”
Counts come from 1,795,640 Hacker News comments, November 2025 to November 2026. How we count. Report a mistake.