KaLM Embedding Multilingual Mini Instruct v2.5 measured on 21 of 261 Embeddings benchmarks21
BenchmarkByScoreRank
Retrieval8
Other13
No benchmark matches
self-host onlyprice
58.5MTEB English v2
Sep 2025released
896Dimensions
33K tokensMax input
en, zhLanguages
Rankingconfiguration unstated?
Full ranking →···
11Qwen 3 Embedding 4B72.312KaLM Embedding Multilingual Mini Instruct v2.570.913Zhinao ChineseModernBert Embedding68.5Strong at
Alternatives?
ModelScoreCostPriceWhat changes
BetterQwen 3 Embedding 8BunstatedOpen+4%—$0.01Languages: not verified+1KOpenQZhou Embedding Zh+11%——Max input: 33K tokens → 8K tokens+2
BetterYoutu EmbeddingOpen+9%——Max input: 33K tokens → 1K tokens+2KBetterQZhou EmbeddingunstatedOpen+9%——Memory (4-bit): 0.3 GB → 4.2 GB
BetterSeed1.5 Embeddingunstated+6%——Languages: not verifiedModelScoreCostPriceWhat changes
Nothing here yet.
ModelScoreCostPriceWhat changes
Qwen 3 Embedding 8BunstatedOpen+4%—$0.01Languages: not verified+1KQZhou Embedding ZhOpen+11%——Max input: 33K tokens → 8K tokens+2Youtu EmbeddingOpen+9%——Max input: 33K tokens → 1K tokens+2KQZhou EmbeddingunstatedOpen+9%——Memory (4-bit): 0.3 GB → 4.2 GBSeed1.5 Embeddingunstated+6%——Languages: not verifiedModelScoreCostPriceWhat changes
KQZhou Embedding ZhOpen+11%——Max input: 33K tokens → 8K tokens+2Youtu EmbeddingOpen+9%——Max input: 33K tokens → 1K tokens+2KQZhou EmbeddingunstatedOpen+9%——Memory (4-bit): 0.3 GB → 4.2 GBQwen 3 Embedding 8BunstatedOpen+4%—$0.01Languages: not verified+1iJasper Token Compression 600MunstatedOpen+4%——Memory (4-bit): 0.3 GB → 0.4 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 KaLM Embedding Multilingual Mini Instruct v2.5 cost?›
There is no per-unit API price: self-host only.
Is KaLM Embedding Multilingual Mini Instruct v2.5 free to use?›
There is no free API. The weights are open, so you can run it yourself.
How good is KaLM Embedding Multilingual Mini Instruct v2.5, and what is it best at?›
#35 of 200 in Embeddings. Strongest at: Embeddings & Retrieval.
Is KaLM Embedding Multilingual Mini Instruct v2.5 better than Qwen 3 Embedding 4B?›
On MTEB Mandarin Chinese v1 · Mean over tasks · Mandarin Chinese, KaLM Embedding Multilingual Mini Instruct v2.5 scores 70.9 vs 72.3 — lower.
When was KaLM Embedding Multilingual Mini Instruct v2.5 released? Is there a newer version?›
Released 2025-09-30. It is the current version in its line.
Is KaLM Embedding Multilingual Mini Instruct v2.5 open source?›
Yes, the weights are open (Apache-2.0 licence).
Can I run KaLM Embedding Multilingual Mini Instruct v2.5 locally?›
The weights are open and can be self-hosted; hardware needs are not confirmed yet.
What is KaLM Embedding Multilingual Mini Instruct v2.5?›
KaLM Embedding Multilingual Mini Instruct v2.5 is an open-weights embeddings & retrieval model from KaLM-Embedding, released Sep 2025.
What are the best alternatives to KaLM Embedding Multilingual Mini Instruct v2.5?›
Qwen 3 Embedding 8B (better), QZhou Embedding Zh (open weights).