OpenLuxJina Reranker V3 API
Jina Reranker V3 takes a list of candidates and puts them in the right order. On OpenLux it costs $0.0042 per million tokens — the smallest line item in a RAG pipeline, and often the one that decides whether the answer is right.
Pay-as-you-go rate
$0.0042 / 1M tokens
Billed in USD, metered per request with real-time cost in the console.
Access
One OpenLux key and the OpenAI-compatible endpoint: set base_url to https://openlux.lol/v1 and pass jina-reranker-v3 as the model.
Get started
Create a key in the console, then follow the five-minute docs for streaming, tools and error codes.
What Jina Reranker V3 delivers
Vector search is fast and broad, but the top results are not always the best ones. A reranker reads the query and each candidate together and scores relevance properly, which is usually the difference between a model that answers from the right passage and one that answers from a plausible-looking neighbour.
The pattern is two-stage retrieval: embed broadly with an embedding model, then rerank the top candidates before generation. Both stages are metered by tokens on OpenLux and both are cheap — a reranker call over twenty candidates costs a tiny fraction of a cent.
Because it shares the endpoint with generation, the whole chain — embed, rerank, answer — is one integration with one key and one bill, and each stage’s cost is visible separately in the console.
Call Jina Reranker V3 in minutes
Every OpenLux model speaks the OpenAI protocol. Reuse your existing SDKs and replace two values: the base URL and the API key.
cURL
curl https://openlux.lol/v1/chat/completions \
-H "Authorization: Bearer $OPENLUX_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "jina-reranker-v3",
"messages": [{"role": "user", "content": "Query: how is OpenLux billed? Candidate passages about pricing, rate limits and credits."}]
}'Python (OpenAI SDK)
from openai import OpenAI
client = OpenAI(
api_key="sk-openlux-...",
base_url="https://openlux.lol/v1",
)
resp = client.chat.completions.create(
model="jina-reranker-v3",
messages=[{"role": "user", "content": "Query: how is OpenLux billed? Candidate passages about pricing, rate limits and credits."}],
)
print(resp.choices[0].message.content)Node.js (OpenAI SDK)
import OpenAI from "openai";
const client = new OpenAI({
apiKey: process.env.OPENLUX_API_KEY,
baseURL: "https://openlux.lol/v1",
});
const resp = await client.chat.completions.create({
model: "jina-reranker-v3",
messages: [{ role: "user", content: "Query: how is OpenLux billed? Candidate passages about pricing, rate limits and credits." }],
});
console.log(resp.choices[0].message.content);Where Jina Reranker V3 fits
Relevance scoring
Scores each candidate against the query in context, rather than by vector distance alone.
Better context windows
Sends the generator fewer, better passages, which improves answers and reduces token spend.
Multilingual input
Works on mixed-language candidate sets, which suits global documentation.
Token metering
At $0.0042 per million tokens, reranking adds negligible cost to a query path.
RAG accuracy
Add a rerank stage between retrieval and generation to raise answer quality without changing the model above.
Search quality
Improve result ordering on a product or internal search page without rebuilding the index first.
Evidence selection
Pick the best supporting passages for summarisation, citations and research tools.
How jina-reranker-v3 is priced on OpenLux
Jina Reranker V3 is metered pay as you go at $0.0042 / 1M tokens. There is no subscription requirement for API access: top up a balance, call the model, and see the cost of every request in the console.
| Model | Provider | Rate | Discount |
|---|---|---|---|
| jina-reranker-v3 (this model) | AIGC | $0.0042 / 1M tokens | — |
| bge-m3 | AIGC | $0.0035 / 1M tokens | — |
| qwen3-max | Qwen | $0.0784 / 1M tokens | -88% |
| deepseek-v4 | DeepSeek | $0.0613 / 1M tokens | -90% |
Rates follow official provider pricing and are shown in USD. See the full model catalogue and pricing plans for credit allowances.
Jina Reranker V3 questions
How much does Jina Reranker V3 cost on OpenLux?
$0.0042 per 1 million tokens, metered pay as you go. In practice, reranking a page of candidates costs a fraction of a cent.
How do I call Jina Reranker V3?
Send "model": "jina-reranker-v3" to https://openlux.lol/v1 with your query and the candidate passages, using your OpenLux API key.
Do I need embeddings as well?
Usually yes: embeddings retrieve a broad candidate set and the reranker orders it. Both models are in the catalogue and billed by tokens on the same account.
Will reranking lower my generation cost?
Often, yes — better-ordered context means fewer passages are needed to answer correctly, which reduces the tokens sent to the generation model.
Start calling Jina Reranker V3 today
Create an OpenLux key and send your first request in minutes. Compare Jina Reranker V3 with the rest of the catalogue on the same endpoint and the same bill.