Embeddings

curl http://localhost:1234/v1/embeddings \
  -H "Content-Type: application/json" \
  -d '{"input": ["The RTX 5090 has 32 GB of memory.", "Cats sleep a lot."], "dimensions": 1024}'

from openai import OpenAI

client = OpenAI(base_url="http://localhost:1234/v1", api_key="not-needed")
docs = ["The RTX 5090 has 32 GB of memory.", "Cats sleep a lot."]
d = client.embeddings.create(model="qwen3-embedding-4b", input=docs).data
q = client.embeddings.create(model="qwen3-embedding-4b", input="How much VRAM does a 5090 have?",
                             extra_body={"input_type": "query"}).data[0].embedding
scores = [sum(a * b for a, b in zip(q, e.embedding)) for e in d]
print(max(zip(scores, docs)))

import OpenAI from "openai";

const client = new OpenAI({ baseURL: "http://localhost:1234/v1", apiKey: "not-needed" });
const r = await client.embeddings.create({ model: "qwen3-embedding-4b", input: ["The RTX 5090 has 32 GB of memory."], dimensions: 1024 });
console.log(r.data[0].embedding.length);   // 1024

{
  "object": "list",
  "data": [
    { "object": "embedding", "index": 0, "embedding": [0.0213, -0.0047, 0.0391, …] },
    { "object": "embedding", "index": 1, "embedding": [-0.0102, 0.0288, 0.0015, …] }
  ],
  "model": "qwen3-embedding-4b",
  "usage": { "prompt_tokens": 17, "total_tokens": 17 },
  "timings": { "ms": 9.6 }
}