Embeddings and Integrations: Embeddings
Overview
Weaviate Embeddings is a service in Weaviate Cloud that simplifies the creation and management of vector embeddings. With Weaviate Embeddings, developers can access to various embedding models without needing to send data to an external provider.
- Fast, flexible development: Simply operations with one less API and vendor to manage. Choose between class-leading OSS and proprietary models.
- Freedom from rate limits: Bring models closer to your data to reduce latency. Enable limitless embeddings per second without artificial constraints.
- GPU-powered and cost-efficient: Maximize performance while managing costs with simple, pay-as-you-go pricing.
Additional Details and Pricing
Snowflake arctic-embed 1.5 = $0.025 per 1M tokens
Snowflake arctic-embed 2.0 = $0.040 per 1M tokens
ModernVBERT colmodernvbert = $0.065 per 1M tokens
Available Models
Text Embedding Models
❄️
Arctic L v2.0Default
Snowflake
- •Multilingual support
- •Up to 8,192 tokens
- •Best for big-scale, complex retrieval
❄️
Arctic M v1.5
Snowflake
- •English support
- •Up to 512 tokens
- •Best for fast, lightweight retrieval
Multimodal Models
🔮
ColModernVBERT
ModernVBERT
- •English support
- •Supports images and document pages
- •Query text limit: 8,092 tokens
- •Best for visual documents (PDFs, slides, invoices) without OCR preprocessing