Embeddings and reranking

Vectors for retrieval and cross-encoder reranking, on the OpenAI embeddings shape and the Cohere and Jina rerank shape.

Base URL https://api.siliconnetwork.io

POST/v1/embeddings
POST/v1/rerank

Embeddings

bge-m3 is the alias of mlx-community/bge-m3-mlx-fp16; qwen3-embedding-0.6b and all-minilm-l6-v2 are the others, and any full id works. The response names the engine id that served. input may be one string, a list of strings, or token id arrays. encoding_format: "base64" packs each vector; dimensions truncates on models that support it.

silicon - zsh

curl -X POST https://api.siliconnetwork.io/v1/embeddings \
  -H "Authorization: Bearer sn_live_..." \
  -H "Content-Type: application/json" \
  -d '{
  "model": "bge-m3",
  "input": [
    "The quick brown fox",
    "jumps over the lazy dog"
  ]
}'

sdk.py

from siliconnetwork import SiliconNetwork

client = SiliconNetwork()
out = client.embeddings.create(**{
  "model": "bge-m3",
  "input": [
    "The quick brown fox",
    "jumps over the lazy dog"
  ]
})
print(len(out.data[0].embedding))

sdk.ts

import { SiliconNetwork } from "@siliconnetwork/sdk";

const client = new SiliconNetwork({ apiKey: process.env.SILICON_API_KEY });
const out = await client.embeddings.create({
  "model": "bge-m3",
  "input": [
    "The quick brown fox",
    "jumps over the lazy dog"
  ]
});
console.log(out.data[0].embedding.length);

response

{
  "object": "list",
  "data": [
    { "object": "embedding", "index": 0, "embedding": [0.0123, -0.0456, "..."] },
    { "object": "embedding", "index": 1, "embedding": [0.0789, 0.0012, "..."] }
  ],
  "model": "mlx-community/bge-m3-mlx-fp16",
  "usage": { "prompt_tokens": 12, "total_tokens": 12 }
}

Responses: 200 Successful Response, 422 Validation Error

POST/v1/rerank

Rerank

silicon - zsh

curl -X POST https://api.siliconnetwork.io/v1/rerank \
  -H "Authorization: Bearer sn_live_..." \
  -H "Content-Type: application/json" \
  -d '{
  "model": "qwen3-reranker-0.6b",
  "query": "How do I rotate an API key?",
  "documents": [
    "Keys are created in the dashboard and shown once.",
    "The relay lets nodes behind NAT join the network.",
    "Revoke a key any time; it stops working within 15 seconds."
  ],
  "top_n": 2,
  "return_documents": true
}'

sdk.py

from siliconnetwork import SiliconNetwork

client = SiliconNetwork()
out = client.rerank.create(**{
  "model": "qwen3-reranker-0.6b",
  "query": "How do I rotate an API key?",
  "documents": [
    "Keys are created in the dashboard and shown once.",
    "The relay lets nodes behind NAT join the network.",
    "Revoke a key any time; it stops working within 15 seconds."
  ],
  "top_n": 2,
  "return_documents": true
})
for r in out.results:
    print(r.index, r.relevance_score)

sdk.ts

import { SiliconNetwork } from "@siliconnetwork/sdk";

const client = new SiliconNetwork({ apiKey: process.env.SILICON_API_KEY });
const out = await client.rerank.create({
  "model": "qwen3-reranker-0.6b",
  "query": "How do I rotate an API key?",
  "documents": [
    "Keys are created in the dashboard and shown once.",
    "The relay lets nodes behind NAT join the network.",
    "Revoke a key any time; it stops working within 15 seconds."
  ],
  "top_n": 2,
  "return_documents": true
});
for (const r of out.results) console.log(r.index, r.relevance_score);

response

{
  "model": "mlx-community/Qwen3-Reranker-0.6B-mxfp8",
  "results": [
    { "index": 2, "relevance_score": 0.91, "document": { "text": "Revoke a key any time; it stops working within 15 seconds." } },
    { "index": 0, "relevance_score": 0.74, "document": { "text": "Keys are created in the dashboard and shown once." } }
  ],
  "usage": { "prompt_tokens": 58, "total_tokens": 58 }
}

Responses: 200 Successful Response, 422 Validation Error

Silicon Network›Docs›API›Embeddings and reranking37 items