Alibaba · reranker
Qwen3 Rerank API
Qwen3 Rerank is Alibaba Cloud Model Studio’s hosted text-reranking API. It reorders candidate documents for search and RAG, with API-specific document and token limits.
Hosted task API
Official sources reviewed 2026-10-02. This page documents access and specifications; it does not report a model benchmark.
Task and deployment contract
Hosted task API; use the exact model ID and the region-specific endpoint in the official documentation.
Hosted API ID: qwen3-rerank. The Singapore interface accepts up to 500 candidate documents.
The documented limits are 4000 tokens per item and 120000 tokens per request. Do not substitute the open-weight checkpoint’s 32K sequence limit.
Return scores to your application using the original document IDs or result indexes. A new relevance threshold needs task-specific validation.
This task uses a reranking endpoint rather than chat completions. Consult the linked pricing documentation for the current region and billing unit.
Billing and limits
No current provider rate for this exact configuration is recorded here. Open weights still require compute and operations; an unlisted rate does not mean a hosted API is free.
Recorded input window: 4,000 tokens. Apply the endpoint's stricter request and batch limits when applicable.
Fast Pivot — Access hundreds of AI models from a single API — competitive pricing, reliable uptime, enterprise-grade infrastructure.