Spectral Embed v1 140M measured on 5 of 261 Embeddings benchmarks5
BenchmarkByScoreRank
No benchmark matches
self-host onlyprice
64.4MTEB English v2
Oct 2026released
1K tokensMax input
enLanguages
Alternatives?
ModelScoreCostPriceWhat changes
ModelScoreCostPriceWhat changes
Nothing here yet.
ModelScoreCostPriceWhat changes
ModelScoreCostPriceWhat changes
cF2LLM 4BunstatedOpen+42%——Memory (4-bit): 0.1 GB → 2.4 GBcF2LLM v2 4BunstatedOpen+42%——Memory (4-bit): 0.1 GB → 2.4 GBaGiga Embeddings Instruct 10B A1.8B 0826unstatedOpen+42%——Memory (4-bit): 0.1 GB → 6.3 GBcF2LLM v2 8BunstatedOpen+42%——Memory (4-bit): 0.1 GB → 4.5 GBaGiga Embeddings Instruct 3B 0826unstatedOpen+42%——Memory (4-bit): 0.1 GB → 1.9 GBModelScoreCostPriceWhat changes
JSpectral Embed v2 140MOpenn/a——Max input: not verified+2All JMullings 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
LicenceMITcommercial use with conditions
Benchmarks
in Embeddings & RetrievalFAQ
How much does Spectral Embed v1 140M cost?›
There is no per-unit API price: self-host only.
Is Spectral Embed v1 140M free to use?›
There is no free API. The weights are open, so you can run it yourself.
How good is Spectral Embed v1 140M, and what is it best at?›
#202 of 222 in Embeddings.
Is Spectral Embed v1 140M better than Static Retrieval Mrl En v1?›
On MTEB English v2 · Classification · English, Spectral Embed v1 140M scores 64.4 vs 64.5 — about the same.
When was Spectral Embed v1 140M released? Is there a newer version?›
Released 2026-10-06. It is the current version in its line.
Is Spectral Embed v1 140M open source?›
Yes, the weights are open (MIT licence).
Can I run Spectral Embed v1 140M locally?›
The weights are open and can be self-hosted; hardware needs are not confirmed yet.
What is Spectral Embed v1 140M?›
Spectral Embed v1 140M is an open-weights embeddings & retrieval model from JMullings, released Oct 2026.
What are the best alternatives to Spectral Embed v1 140M?›
EmbeddingGemma 300M (better), F2LLM 4B (open weights).