Tarka Embedding 150M v1 measured on 8 of 194 Embeddings benchmarks8
BenchmarkByScoreRank
No benchmark matches
self-host onlyprice
51.5MTEB English v2
Nov 2025released
2K tokensMax input
en, es, fr +5Languages
Rankingconfiguration unstated?
Full ranking →···
92Mini Gte81.693Tarka Embedding 150M v181.594Bilingual Embedding Small81.5Strong at
Alternatives?
ModelScoreCostPriceWhat changes
ModelScoreCostPriceWhat changes
Nothing here yet.
ModelScoreCostPriceWhat changes
ModelScoreCostPriceWhat changes
KQZhou EmbeddingunstatedOpen+12%——Languages: loses es, fr, de +3+1iJasper Token Compression 600MunstatedOpen+9%——Languages: loses es, fr, de +3+1ModelScoreCostPriceWhat changes
TTarka Embedding 350M v1unstatedOpen+4%——Memory (4-bit): 0.1 GB → 0.2 GBAll Tarka-AIR models →Run it yourself
estimateFits on✓ 8 GB (phone, laptop) · full quality✓ 16 GB · full quality✓ 24 GB GPU · full quality✓ 48 GB GPU · full quality✓ 80 GB GPU · full quality
LicenceGemma termscommercial use with conditions
Benchmarks
in Embeddings & RetrievalFAQ
How much does Tarka Embedding 150M v1 cost?›
There is no per-unit API price: self-host only.
Is Tarka Embedding 150M v1 free to use?›
There is no free API. The weights are open, so you can run it yourself.
How good is Tarka Embedding 150M v1, and what is it best at?›
#95 of 200 in Embeddings.
Is Tarka Embedding 150M v1 better than Mini Gte?›
On MTEB English v2 · STS · English, Tarka Embedding 150M v1 scores 81.5 vs 81.6 — about the same.
When was Tarka Embedding 150M v1 released? Is there a newer version?›
Released 2025-11-04. It is the current version in its line.
Is Tarka Embedding 150M v1 open source?›
Yes, the weights are open (gemma licence).
Can I run Tarka Embedding 150M v1 locally?›
The weights are open and can be self-hosted; hardware needs are not confirmed yet.
What is Tarka Embedding 150M v1?›
Tarka Embedding 150M v1 is an open-weights embeddings & retrieval model from Tarka-AIR, released Nov 2025.
What are the best alternatives to Tarka Embedding 150M v1?›
EmbeddingGemma 300M (better), QZhou Embedding (open weights).