All models
Model card

BGE Large EN v1.5

sdn-bge-large-enEmbeddingsLive in the catalog

The highest-accuracy English option: 1024 dimensions give it room to separate passages that a smaller model collapses together. Reach for it when retrieval quality is the bottleneck in your pipeline and a wider index is an acceptable cost.

Max input1K tokens
Input · per Mtok$0.21
Dimensions1,024
AvailabilityLive
Where it earns its keep[01/03]
  • 01Best English retrieval accuracy here
  • 021024 dimensions for finer separation
  • 03Strong on near-duplicate passages
Capabilities
EmbeddingsYes

Turns text into a fixed-length vector for semantic search, RAG, clustering, and dedupe. Call it on /v1/embeddings with a string or array of inputs — no chat, tools, or reasoning.

Behind the endpoint[02/03]

One endpoint. Served by our engine.

BGE Large EN v1.5 is served through the Sideren engine — zero-downtime serving is the design target, not a status-page apology. You request it by name; everything else is our problem.

Call it by name
curl https://api.sideren.io/v1/embeddings \
  -H "authorization: Bearer sdn_your_key" \
  -H "content-type: application/json" \
  -d '{
    "model": "sdn-bge-large-en",
    "input": "text to embed"
  }'

OpenAI-style clients work too — POST the same model name to /v1/embeddings with a Bearer key. See the docs for both dialects.

Put your agent on real infrastructure

Your agent doesn't change.
Everything underneath does.

$export ANTHROPIC_BASE_URL=https://api.sideren.io

Start free · no card · Starter from $5/mo