Ko Embed v0 measured on 8 of 194 Embeddings benchmarks8
BenchmarkByScoreRank
No benchmark matches
self-host onlyprice
58.5MTEB Korean v2
Sep 2026released
768Dimensions
16K tokensMax input
en, koLanguages
Alternatives?
ModelScoreCostPriceWhat changes
ModelScoreCostPriceWhat changes
Nothing here yet.
ModelScoreCostPriceWhat changes
ModelScoreCostPriceWhat changes
nKURE v2Open+36%——Max input: 16K tokens → 8K tokensnKURE v3Open+35%——Max input: 16K tokens → 8K tokens+1nKURE Sparse BaseOpen+34%——Max input: 16K tokens → 8K tokens+1sComsat Embed Ko 8B PreviewOpen+33%——Memory (4-bit): 0.1 GB → 4.5 GBRun 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 & RetrievalFAQ
How much does Ko Embed v0 cost?›
There is no per-unit API price: self-host only.
Is Ko Embed v0 free to use?›
There is no free API. The weights are open, so you can run it yourself.
How good is Ko Embed v0, and what is it best at?›
#30 of 156 in Embeddings.
Is Ko Embed v0 better than Granite Embedding 278M Multilingual?›
On MTEB Korean v2 · Retrieval · multilingual (2 languages), Ko Embed v0 scores 59.6 vs 61.2 — lower.
When was Ko Embed v0 released? Is there a newer version?›
Released 2026-09-10. It is the current version in its line.
Is Ko Embed v0 open source?›
Yes, the weights are open (Apache-2.0 licence).
Can I run Ko Embed v0 locally?›
The weights are open and can be self-hosted; hardware needs are not confirmed yet.
What is Ko Embed v0?›
Ko Embed v0 is an open-weights embeddings & retrieval model from NGA-KR, released Sep 2026.
What are the best alternatives to Ko Embed v0?›
Qwen 3 Embedding 8B (better), KURE v2 (open weights).