Models / All Embed Large v3

All Embed Large v3

allais/embed-large-v3

High-recall embedding model for RAG and semantic search.

embeddingembeddingsretrieval
Parameters
7B
Context window
33K tokens
Serverless $/1M
$0.1753 in · $0 out
reference $0.08 · $0
Dedicated
1× L40S-48G
sourced live, undercuts $1.9/hr

Use via API

curl http://localhost:3000/api/v1/chat/completions \
  -H "Authorization: Bearer $ALLAIS_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model": "allais/embed-large-v3", "messages": [{"role": "user", "content": "Hello"}]}'
Try in playgroundDeploy on dedicated GPUs