all-MiniLM-L6-v2
Ranked 7 of 8 embedding models by recommenders
A small, free, open embedding model that runs on-device and gives good semantic matching for its size. Critics say it is old and behind newer open-weight models.
A small, free, open embedding model that runs on-device and gives good semantic matching for its size. Critics say it is old and behind newer open-weight models.
9 people gave an opinion
- recommend it
- 6
- criticize it
- 3
- approval share
- 67%
Recommended in 6 of 13 months
- November 2025: recommended
- December 2025: recommended
- January 2026: recommended
- February 2026: not recommended
- March 2026: recommended
- April 2026: not recommended
- May 2026: recommended
- June 2026: not recommended
- July 2026: not 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
- 6 people recommend itNeeds 10 or more, not met.
- 2 recommenders for every criticNeeds 2 to 1 or better, met.
- Recommended in 6 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
Small enough to run on-device
“embedding models we had for years like minlm (80mb) are so small, and I really wonder why not more on device searches use something like it”
Comment from 6 Dec 2025 in “Why are your models so big? (2023)” “One thing that's still compelling about all-Mini is that it's feasible to use it client-side.”
Comment from 28 Nov 2025 in “28M Hacker News comments as vector embedding search dataset” Good semantic matching for its size
“Works well enough for semantic matching against ~50k transaction descriptions.”
Comment from 15 Jan 2026 in “Ask HN: How are you doing RAG locally?” “only 1.3 points behind bge-large at 15x smaller”
Comment from 24 Sep 2026 in “Kev: Tiny Jev-like family of decision models built on top of Qwen3.5”
What people warn about
Each mark is one comment
Outdated compared with newer models
“it's old and does not implement the newest advances in architectures and data training pipelines”
Comment from 28 Nov 2025 in “28M Hacker News comments as vector embedding search dataset” “This is so old and terribly behind the new models”
Comment from 28 Nov 2025 in “28M Hacker News comments as vector embedding search dataset” “there's much better open weights models out there”
Comment from 28 Nov 2025 in “28M Hacker News comments as vector embedding search dataset”
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 | — | — |
| all-MiniLM-L6-v2This page | 6 | 3 | 67% | 6 | — | — |
What people said
8 of the 10 comments that mention it, newest first. Each links to the original comment.
“only 1.3 points behind bge-large at 15x smaller”
“For a fast, open, and local model, I've found it hard to beat”
“Works well enough for semantic matching against ~50k transaction descriptions.”
“embedding models we had for years like minlm (80mb) are so small, and I really wonder why not more on device searches use something like it”
“One thing that's still compelling about all-Mini is that it's feasible to use it client-side.”
“there's much better open weights models out there”
“This is so old and terribly behind the new models”
“it's old and does not implement the newest advances in architectures and data training pipelines”
Counts come from 1,795,640 Hacker News comments, November 2025 to November 2026. How we count. Report a mistake.