Bekko Embedding is an ultra-compact (AP 8M), high-performance multilingual model. Fast on CPUs, with 8K　token inputs and 384/256/128/64D outputs.

embedding

ollama pull hotchpotch/bekko-embedding-v1-a8m:q8_0

curl http://localhost:11434/api/embed \
  -d '{
    "model": "hotchpotch/bekko-embedding-v1-a8m:q8_0",
    "input": "Why is the sky blue?"
  }'

import ollama

response = ollama.embed(
    model='hotchpotch/bekko-embedding-v1-a8m:q8_0',
    input='The sky is blue because of Rayleigh scattering',
)
print(response.embeddings)

import ollama from 'ollama'

const response = await ollama.embed({
  model: 'hotchpotch/bekko-embedding-v1-a8m:q8_0',
  input: 'The sky is blue because of Rayleigh scattering',
})
console.log(response.embeddings)

Details

Updated 2 months ago

2 months ago

3978227bd8ff · 119MB

model

archmodern-bert

parameters106M

quantizationQ8_0

119MB

params

{ "num_ctx": 8192 }

17B

Readme

bekko-embedding-v1-a8m

A tiny multilingual embedding model with 384-dimensional output and an 8192-token context.

Choose the tag for your machine:

GPU or Apple Silicon: hotchpotch/bekko-embedding-v1-a8m (BF16, default)
CPU: hotchpotch/bekko-embedding-v1-a8m:q8_0
Explicit FP16: hotchpotch/bekko-embedding-v1-a8m:f16

ollama pull hotchpotch/bekko-embedding-v1-a8m

curl http://localhost:11434/api/embed \
  -d '{"model":"hotchpotch/bekko-embedding-v1-a8m","input":"What is the tallest mountain in Japan?"}'

Model details and other runtimes: https://huggingface.co/hotchpotch/bekko-embedding-v1-a8m

Bekko Embedding is an ultra-compact (AP 8M), high-performance multilingual model. Fast on CPUs, with 8K token inputs and 384/256/128/64D outputs.

Details

Readme

bekko-embedding-v1-a8m

Bekko Embedding is an ultra-compact (AP 8M), high-performance multilingual model. Fast on CPUs, with 8K　token inputs and 384/256/128/64D outputs.