LateOn Code Edge Pretrain measured on 6 of 261 Embeddings benchmarks6
BenchmarkByScoreRank
No benchmark matches
self-host onlyprice
54CoIR
Feb 2026released
2K tokensMax input
enLanguages
Alternatives?
ModelScoreCostPriceWhat changes
BetterEmbeddingGemma 300MunstatedOpen+25%—$0.002Languages: not verified+1
OpenQwen 3 Embedding 8Bunstated+48%—$0.01Languages: not verified+1
Betterpplx Embed v1 0.6BOpen+38%—$0.004Languages: not verified+1BetterQwen 3 Embedding 0.6BunstatedOpen+39%—$0.01Languages: not verified+1BetterQwen 3 Embedding 4BunstatedOpen+47%—$0.02Languages: not verified+1ModelScoreCostPriceWhat changes
Nothing here yet.
ModelScoreCostPriceWhat changes
EmbeddingGemma 300MunstatedOpen+25%—$0.002Languages: not verified+1pplx Embed v1 0.6BOpen+38%—$0.004Languages: not verified+1Qwen 3 Embedding 8BunstatedOpen+48%—$0.01Languages: not verified+1Qwen 3 Embedding 0.6BunstatedOpen+39%—$0.01Languages: not verified+1Qwen 3 Embedding 4BunstatedOpen+47%—$0.02Languages: not verified+1ModelScoreCostPriceWhat changes
Qwen 3 Embedding 8BunstatedOpen+48%—$0.01Languages: not verified+1cC2LLM 7BOpen+48%——Memory (4-bit): 0 GB → 4.6 GBcF2LLM v2 8BunstatedOpen+47%——Memory (4-bit): 0 GB → 4.5 GBcF2LLM v2 4BunstatedOpen+47%——Memory (4-bit): 0 GB → 2.4 GBQwen 3 Embedding 4BunstatedOpen+47%—$0.02Languages: not verified+1ModelScoreCostPriceWhat changes
lLateOn CodeOpen+36%——Memory (4-bit): 0 GB → 0.1 GBlmLateOnOpen+32%——Memory (4-bit): 0 GB → 0.2 GBlmDenseOnOpen+28%——Memory (4-bit): 0 GB → 0.2 GBlLateOn Code EdgeOpen+20%——lLateOn Code PretrainOpen+12%——Memory (4-bit): 0 GB → 0.1 GBlDenseOnOpenn/a——Max input: 2K tokens → 1K tokens+1lDenseOn UnsupervisedOpenn/a——Max input: 2K tokens → 1K tokens+1lLateOnOpenn/a——Max input: 2K tokens → 0K tokens+1All lightonai 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 & RetrievalFAQ
How much does LateOn Code Edge Pretrain cost?›
There is no per-unit API price: self-host only.
Is LateOn Code Edge Pretrain free to use?›
There is no free API. The weights are open, so you can run it yourself.
How good is LateOn Code Edge Pretrain, and what is it best at?›
#46 of 58 in Embeddings.
Is LateOn Code Edge Pretrain better than Bekko Embedding v1 A8m?›
On CoIR · Mean over tasks · multilingual (9 languages), LateOn Code Edge Pretrain scores 54 vs 54.2 — about the same.
When was LateOn Code Edge Pretrain released? Is there a newer version?›
Released 2026-02-12. It is the current version in its line.
Is LateOn Code Edge Pretrain open source?›
Yes, the weights are open (Apache-2.0 licence).
Can I run LateOn Code Edge Pretrain locally?›
The weights are open and can be self-hosted; hardware needs are not confirmed yet.
What is LateOn Code Edge Pretrain?›
LateOn Code Edge Pretrain is an open-weights embeddings & retrieval model from lightonai, released Feb 2026.
What are the best alternatives to LateOn Code Edge Pretrain?›
EmbeddingGemma 300M (better), Qwen 3 Embedding 8B (open weights).