Tarka Embedding 350M v1 measured on 8 of 261 Embeddings benchmarks8
BenchmarkByScoreRank
No benchmark matches
self-host onlyprice
55.1MTEB English v2
Nov 2025released
1,024Dimensions
128K tokensMax input
en, es, fr +5Languages
Rankingconfiguration unstated?
Full ranking →···
62Mxbai Embed Large v155.463Tarka Embedding 350M v155.164BidirLM Omni 2.5B Embedding55.1Strong at
Alternatives?
ModelScoreCostPriceWhat changes
BetterQwen 3 Embedding 8BunstatedOpen+26%—$0.01Max input: 128K tokens → 33K tokens+2IOpenYuan Embedding 2.0 Enunstated+28%——Max input: 128K tokens → 33K tokens+2BetterQwen 3 Embedding 0.6BunstatedOpen+12%—$0.01Max input: 128K tokens → 33K tokens+2BetterQwen 3 Embedding 4BunstatedOpen+24%—$0.02Max input: 128K tokens → 33K tokens+2
BetterJina Embeddings v5 Text NanounstatedOpen+7%—$0.02Max input: 128K tokens → 8K tokens+1ModelScoreCostPriceWhat changes
Nothing here yet.
ModelScoreCostPriceWhat changes
Qwen 3 Embedding 8BunstatedOpen+26%—$0.01Max input: 128K tokens → 33K tokens+2Qwen 3 Embedding 0.6BunstatedOpen+12%—$0.01Max input: 128K tokens → 33K tokens+2Qwen 3 Embedding 4BunstatedOpen+24%—$0.02Max input: 128K tokens → 33K tokens+2Jina Embeddings v5 Text NanounstatedOpen+7%—$0.02Max input: 128K tokens → 8K tokens+1iE5 Mistral 7B InstructunstatedOpen+4%—$0.022Max input: 128K tokens → 33K tokens+2ModelScoreCostPriceWhat changes
IYuan Embedding 2.0 EnunstatedOpen+28%——Max input: 128K tokens → 33K tokens+2Qwen 3 Embedding 8BunstatedOpen+26%—$0.01Max input: 128K tokens → 33K tokens+2Qwen 3 Embedding 4BunstatedOpen+24%—$0.02Max input: 128K tokens → 33K tokens+2KQZhou EmbeddingunstatedOpen+22%——Max input: 128K tokens → 33K tokens+2iJasper Token Compression 600MunstatedOpen+20%——Max input: 128K tokens → 33K tokens+2ModelScoreCostPriceWhat changes
TTarka Embedding 150M v1unstatedOpen-7%——Max input: 128K tokens → 2K tokensAll 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
Licencelfm1.0commercial use with conditions
Benchmarks
in Embeddings & RetrievalFAQ
How much does Tarka Embedding 350M v1 cost?›
There is no per-unit API price: self-host only.
Is Tarka Embedding 350M v1 free to use?›
There is no free API. The weights are open, so you can run it yourself.
How good is Tarka Embedding 350M v1, and what is it best at?›
#59 of 200 in Embeddings.
Is Tarka Embedding 350M v1 better than Mxbai Embed Large v1?›
On MTEB English v2 · Retrieval · English, Tarka Embedding 350M v1 scores 55.1 vs 55.4 — about the same.
When was Tarka Embedding 350M v1 released? Is there a newer version?›
Released 2025-11-11. It is the current version in its line.
Is Tarka Embedding 350M v1 open source?›
Yes, the weights are open (lfm1.0 licence).
Can I run Tarka Embedding 350M v1 locally?›
The weights are open and can be self-hosted; hardware needs are not confirmed yet.
What is Tarka Embedding 350M v1?›
Tarka Embedding 350M v1 is an open-weights embeddings & retrieval model from Tarka-AIR, released Nov 2025.
What are the best alternatives to Tarka Embedding 350M v1?›
Qwen 3 Embedding 8B (better), Yuan Embedding 2.0 En (open weights).