Home › Models › Qwen3 Embedding 8B

Qwen3 Embedding 8B API

Near-zero-cost embeddings for retrieval and search — one OpenAI-compatible call, metered on input tokens only.

EmbeddingBest for: Near-zero-cost embeddings for RAG and search.

Why developers pick it

• One OpenAI-compatible /v1/embeddings call — no new SDK.
• Metered per input token; audit trail included like every Bridge request.

Code example

from openai import OpenAI

client = OpenAI(base_url="https://bridge.computeindex.net/v1", api_key="sk-bridge-YOUR_KEY")

emb = client.embeddings.create(model="qwen3-embedding-8b", input="Bridge: one API for Chinese frontier models")
print(len(emb.data[0].embedding))

FAQ

Which endpoint? Standard /v1/embeddings with model="qwen3-embedding-8b".

How is it billed? Input tokens only, $0.02 per 1M.

Related