Alibaba Cloud: Qwen3.7-Text-Rerank
qwen3.7-text-rerank- Input / Output
- $0.0735 / $0.00
- Context
- 120K
- Modalities
- Reranking
About Qwen3.7-Text-Rerank
Qwen3.7-Text-Rerank is Alibaba's text-only reranker. It scores each candidate document against a query and returns them reordered, so a search or RAG system can pass only the most relevant passages onward. It covers 201 languages and dialects and takes many documents per request, with long items allowed. It is the pick for text retrieval; Qwen3-VL-Rerank is the one for image and video candidates.
Where it works well
- Reorders retrieved passages by relevance to the query before they reach an answering model.
- Supports 201 languages and dialects, a very wide range for multilingual search.
- Accepts up to 500 documents per request, enough to rerank a broad first-stage shortlist.
- Individual items can be long, which suits full paragraphs and articles.
When to choose another model
- It reads text only; images and video need Qwen3-VL-Rerank.
- It does not retrieve anything: you must supply the candidate documents from your own search.
- It returns scores and ordering, not generated answers or citations.
Getting started
Create API key
Create a key in Console, then use it with every model on the platform.
Send your first request
Copy the example for your language and run it against the endpoint.
RerankingTokenLab endpointPOST/v1/rerankcurl -X POST "https://api.tokenlab.sh/v1/rerank" \ -H "Authorization: Bearer sk-xxx" \ -H "Content-Type: application/json" \ -d '{ "operation": "rerank", "model": "qwen3.7-text-rerank", "query": "What is the capital of France?", "documents": [ "Paris is the capital of France.", "Berlin is the capital of Germany." ] }'
Pricing
The Verified price applies to Verified, which costs less on most models. The Official price is the model maker's published price and applies to the more reliable Official route. Auto bills the route that completes the request.
Pricing
per 1M tokens- Verified price
- —
- Official price
- Input$0.0735per 1M tokensOutput$0.00per 1M tokens
| Official priceper 1M tokens | Verified priceper 1M tokens | Discount | |
|---|---|---|---|
| Input | $0.0735 | — | — |
| Output | $0.00 | — | — |
Usage & activity
Success rate is the share of requests that completed. Latency is how long a full response takes; P95 means 95% of requests finished within that time.
Usage & availability
Last 24 hours- Requests
- Success rate
- P95 latency
- Total tokens
Data is based on aggregate user requests, excluding status checks.
Open in Console
Open Qwen3.7-Text-Rerank in Console with a prompt ready to edit or send.
Help me try qwen3.7-text-rerank with rerank at /v1/rerank. Show the result and cost.
Use cases
RAG answer quality
Rerank vector-search hits for a question so the best few passages go into the prompt.
Multilingual site search
Order results for queries written in one language against documents in another.
Support article matching
Pick the knowledge-base article that best fits an incoming ticket from a shortlist of candidates.
Prompt examples
Query: how do I cancel an order after it ships. Rank these 40 help-center paragraphs.
Instruction: prefer official policy over forum posts. Query: refund window for annual plans.
Query in Indonesian: cara mengganti kata sandi. Rank these English documents.
Cost calculator
FAQ
What is a reranker for?
It refines search results. A first-stage search returns many loose matches; the reranker reads each against the query and orders them by relevance, which usually improves what an LLM sees in retrieval-augmented generation.
How many documents can I send?
Up to 500 documents per request. If your first stage returns more, cut the shortlist to the top 500 before reranking. Candidates must be text, since images and video go to Qwen3-VL-Rerank.
Which languages does it cover?
Alibaba lists 201 languages and dialects, including Chinese, English, Spanish, French, Portuguese, Indonesian, Japanese, Korean, German, and Russian. Cross-language ranking, such as a query in one language over documents in another, is a common use.
Should I use this or Qwen3-VL-Rerank?
Use this one when candidates are plain text. Qwen3-VL-Rerank is needed when images or videos are part of the candidates or query, but supports fewer languages and fewer documents.
Can I customize relevance?
Alibaba's Qwen3 rerank models accept an instruction that describes the ranking goal. For example, it can prefer official policy over forum posts. Test it on your own data.
How much does Qwen3.7-Text-Rerank cost?
On TokenLab, Qwen3.7-Text-Rerank costs Input $0.0735 / Output $0.00 per 1M tokens. The pricing table above shows the full breakdown.
Which endpoint should Qwen3.7-Text-Rerank use?
Use https://api.tokenlab.sh/v1/rerank for Qwen3.7-Text-Rerank. The request example below shows the matching code shape.
Which operations does Qwen3.7-Text-Rerank support?
Qwen3.7-Text-Rerank supports Reranking. Select an operation above to see its endpoint and request example.
Compare Qwen3.7-Text-Rerank
Sources
Reviewed Oct 2, 2026
