Ruri v3 70M measured on 7 of 261 Embeddings benchmarks7
BenchmarkByScoreRank
No benchmark matches
self-host onlyprice
79.5JMTEB v2
Apr 2025released
384Dimensions
8K tokensMax input
jaLanguages
Alternatives?
ModelScoreCostPriceWhat changes
BetterEmbeddingGemma 300MunstatedOpen+5%—$0.002Max input: 8K tokens → 2K tokens+2cOpenF2LLM v2 8Bunstated+41%——Memory (4-bit): 0 GB → 4.5 GBiBetterMultilingual E5 Large InstructunstatedOpen+8%—$0.01Max input: 8K tokens → 1K tokens+1iBetterE5 Mistral 7B InstructunstatedOpen+13%—$0.022Memory (4-bit): 0 GB → 4.3 GBcBetterF2LLM v2 14BunstatedOpen+43%——Memory (4-bit): 0 GB → 8.4 GBModelScoreCostPriceWhat changes
Nothing here yet.
ModelScoreCostPriceWhat changes
EmbeddingGemma 300MunstatedOpen+5%—$0.002Max input: 8K tokens → 2K tokens+2iMultilingual E5 Large InstructunstatedOpen+8%—$0.01Max input: 8K tokens → 1K tokens+1iE5 Mistral 7B InstructunstatedOpen+13%—$0.022Memory (4-bit): 0 GB → 4.3 GBcF2LLM v2 14BunstatedOpen+43%——Memory (4-bit): 0 GB → 8.4 GBcF2LLM v2 8BunstatedOpen+41%——Memory (4-bit): 0 GB → 4.5 GBModelScoreCostPriceWhat changes
cF2LLM v2 8BunstatedOpen+41%——Memory (4-bit): 0 GB → 4.5 GBcF2LLM v2 4BunstatedOpen+40%——Memory (4-bit): 0 GB → 2.4 GBcF2LLM v2 1.7BunstatedOpen+38%——Memory (4-bit): 0 GB → 1 GBcF2LLM v2 0.6BunstatedOpen+35%——Memory (4-bit): 0 GB → 0.4 GBcF2LLM v2 330MunstatedOpen+30%——Memory (4-bit): 0 GB → 0.2 GBModelScoreCostPriceWhat changes
cRuri v3 310MOpen+4%——Memory (4-bit): 0 GB → 0.2 GBcRuri v3 130MOpen+3%——Memory (4-bit): 0 GB → 0.1 GBcRuri BaseOpensame——Max input: 8K tokens → 1K tokens+1cRuri Base v2Opensame——Max input: 8K tokens → 1K tokens+1cRuri v3 30MOpensame——cRuri LargeOpen-3%——Max input: 8K tokens → 1K tokens+1cRuri Large v2Open-5%——Max input: 8K tokens → 1K tokens+1All cl-nagoya 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
LicenceApache-2.0free for commercial use
Benchmarks
in Embeddings & RetrievalJMTEB v2 · STS · Japanese79.5#47
FAQ
How much does Ruri v3 70M cost?›
There is no per-unit API price: self-host only.
Is Ruri v3 70M free to use?›
There is no free API. The weights are open, so you can run it yourself.
How good is Ruri v3 70M, and what is it best at?›
#47 of 164 in Embeddings.
Is Ruri v3 70M better than Voyage Large 2 Instruct?›
On JMTEB v2 · Clustering · Japanese, Ruri v3 70M scores 50.4 vs 50.5 — about the same.
When was Ruri v3 70M released? Is there a newer version?›
Released 2025-04-09. It is the current version in its line.
Is Ruri v3 70M open source?›
Yes, the weights are open (Apache-2.0 licence).
Can I run Ruri v3 70M locally?›
The weights are open and can be self-hosted; hardware needs are not confirmed yet.
What is Ruri v3 70M?›
Ruri v3 70M is an open-weights embeddings & retrieval model from cl-nagoya, released Apr 2025.
What are the best alternatives to Ruri v3 70M?›
EmbeddingGemma 300M (better), F2LLM v2 8B (open weights).