Nebius Embeddings
Configure Nebius multilingual embedding models
Configuration
embedding_model:
provider: "nebius"
api_key: $NEBIUS_API_KEY
tokenizer: "BAAI/bge-multilingual-gemma2"
model: "bge-multilingual-gemma2"
base_url: "https://api.studio.nebius.ai/v1/"Configuration Options
| Field | Type | Default | Required | Description |
|---|---|---|---|---|
provider | string | - | ✅ | Must be "nebius" |
api_key | string | - | ✅ | Nebius API key |
tokenizer | string | "BAAI/bge-multilingual-gemma2" | ❌ | HuggingFace tokenizer |
model | string | "bge-multilingual-gemma2" | ❌ | Model name |
base_url | string | "https://api.studio.nebius.ai/v1/" | ❌ | Nebius API endpoint |
rpm | integer | 3000 | ❌ | Requests per minute |
tpm | integer | 1000000 | ❌ | Tokens per minute |
Setup
- Create Nebius account: studio.nebius.ai
- Generate API key: API Keys
- Add to environment:
NEBIUS_API_KEY=your-key