Supported Embedding Models
| Provider | Model | Dimensions | Cost / Deployment |
|---|---|---|---|
| OpenAI | text-embedding-3-small | 1536 | Cloud API (Fast & cost-efficient) |
| OpenAI | text-embedding-3-large | 3072 | Cloud API (Maximum accuracy) |
| FastEmbed | BAAI/bge-small-en-v1.5 | 384 | Local ONNX runtime (Zero API cost) |
| Cohere | embed-english-v3.0 | 1024 | Cloud API (High quality search) |
| Ollama | nomic-embed-text | 768 | Local Ollama instance |
Zero-Cost Local FastEmbed
fastembed_config.py
from negentro import Memory
m = Memory.from_config({
"embedder": {
"provider": "fastembed",
"config": {
"model": "BAAI/bge-small-en-v1.5"
}
}
})