Ruri v3 30M measured on 7 of 194 Embeddings benchmarks7
BenchmarkByScoreRank
No benchmark matches
self-host onlyprice
82.2JMTEB v2
Apr 2025released
256Dimensions
8K tokensMax input
jaLanguages
Alternatives?
ModelScoreCostPriceWhat changes
BetterEmbeddingGemma 300MunstatedOpen+3%—$0.002Max input: 8K tokens → 2K tokens+2
OpenQwen 3 Embedding 8Bunstated+7%—$0.01Languages: not verified+1BetterQwen 3 Embedding 4BunstatedOpen+6%—$0.02Languages: not verified+1NBetterJa Embed StsOpen+4%——Max input: 8K tokens → 1K tokens+1cBetterF2LLM v2 8BunstatedOpen+4%——Memory (4-bit): 0 GB → 4.5 GBModelScoreCostPriceWhat changes
Nothing here yet.
ModelScoreCostPriceWhat changes
EmbeddingGemma 300MunstatedOpen+3%—$0.002Max input: 8K tokens → 2K tokens+2Qwen 3 Embedding 8BunstatedOpen+7%—$0.01Languages: not verified+1Qwen 3 Embedding 4BunstatedOpen+6%—$0.02Languages: not verified+1NJa Embed StsOpen+4%——Max input: 8K tokens → 1K tokens+1cF2LLM v2 8BunstatedOpen+4%——Memory (4-bit): 0 GB → 4.5 GBModelScoreCostPriceWhat changes
Qwen 3 Embedding 8BunstatedOpen+7%—$0.01Languages: not verified+1Qwen 3 Embedding 4BunstatedOpen+6%—$0.02Languages: not verified+1NJa Embed StsOpen+4%——Max input: 8K tokens → 1K tokens+1cF2LLM v2 8BunstatedOpen+4%——Memory (4-bit): 0 GB → 4.5 GB
Harrier OSS v1 0.6BunstatedOpen+3%——Memory (4-bit): 0 GB → 0.4 GBModelScoreCostPriceWhat changes
cRuri Large v2Open+2%——Max input: 8K tokens → 1K tokens+1cRuri LargeOpen+2%——Max input: 8K tokens → 1K tokens+1cRuri BaseOpen+1%——Max input: 8K tokens → 1K tokens+1cRuri Base v2Opensame——Max input: 8K tokens → 1K tokens+1cRuri v3 310MOpen-2%——Memory (4-bit): 0 GB → 0.2 GBcRuri v3 70MOpen-3%——cRuri v3 130MOpen-4%——Memory (4-bit): 0 GB → 0.1 GBAll 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 · Japanese82.2#32
FAQ
How much does Ruri v3 30M cost?›
There is no per-unit API price: self-host only.
Is Ruri v3 30M free to use?›
There is no free API. The weights are open, so you can run it yourself.
How good is Ruri v3 30M, and what is it best at?›
#30 of 168 in Embeddings.
Is Ruri v3 30M better than Jina Embeddings v5 Omni Nano?›
On JMTEB v2 · STS · Japanese, Ruri v3 30M scores 82.2 vs 82.2 — about the same.
When was Ruri v3 30M released? Is there a newer version?›
Released 2025-04-07. It is the current version in its line.
Is Ruri v3 30M open source?›
Yes, the weights are open (Apache-2.0 licence).
Can I run Ruri v3 30M locally?›
The weights are open and can be self-hosted; hardware needs are not confirmed yet.
What is Ruri v3 30M?›
Ruri v3 30M is an open-weights embeddings & retrieval model from cl-nagoya, released Apr 2025.
What are the best alternatives to Ruri v3 30M?›
EmbeddingGemma 300M (better), Qwen 3 Embedding 8B (open weights).