qwen3-vl-rerankCompare qwen3-vl-rerank API pricing, supported endpoints, capabilities and access options on Modelsell.
Qwen tokenizer (tiktoken-compat)Scores on standardized evaluations. Higher percentages are better — and rank percentile shows
Metrics sourced fromArtificial Analysis 2026-09-30·Qwen3 Vl Rerank
This model is not included in the current benchmark snapshot.
Missing models or measurements are not zero scores.
Benchmark charts preserve the source model selection and reasoning settings. Missing models or measurements are not zero scores, and benchmark cost or speed is not this site’s service commitment.
Qwen3-VL-Rerank 适合接在关键词或向量检索之后,重新判断查询与每个候选的相关程度。例如搜索“白色、有盖、带把手的陶瓷杯”,初次召回可能混入白色玻璃杯、无盖马克杯和带盖保温杯。重排序会结合查询条件与候选内容打分,帮助筛选更符合要求的结果。
它支持文字或图片作为查询,候选可以是文字、图片和视频。商品图搜、设计参考图筛选、视频素材精排,以及带图片的知识库检索,都可以使用这一流程。图片查询适合从一张参考图中寻找相近内容,文字查询则便于描述主体、动作和场景要求。
先从完整资料库召回一批候选,再把查询和这些候选交给模型,使用 top_n 控制返回数量。需要强调特定排序目标时,可通过 instruct 说明任务,例如优先考虑主体与动作是否匹配。应用应保存候选 ID 与输入顺序,再按返回的 index 找回原始素材。
返回结果包含相关性分数和排序,不会生成向量或聊天答案。分数用于当前请求内比较,不宜把不同查询的分数当作统一的合格线;也不能从未召回的资料中补出结果。视频按片段保存并带上时间范围,能够让搜索结果直接跳转到对应内容。
例如搜索产品照片,在 input.query 中传入 {"text":"白色、有盖、带把手的陶瓷杯"},在 input.documents 中依次放入候选图片对象 {"image":"候选图片URL"},并设置 parameters.top_n=3。
读取 output.results,用每项的 index 对应商品 ID,按 relevance_score 返回排序后的素材。图片可以使用 URL 或 Base64,视频候选使用可公开访问的 URL。
先用人工标注的小组查询比较精排前后前几名的相关性。若目标商品根本没被初次召回,应调整检索或候选数量;若存在颜色、材质等硬性条件,可结合商品字段过滤,再交给模型排序。
Use the linked documentation for model-specific request parameters and examples.
All requests must include Authorization: Bearer <TOKEN> header. Anthropic-formatted endpoints accept the x-api-key header instead.
Generate tokens from the Tokens page; you can scope them to specific models, groups, IPs, and rate-limits.
| Supplier | RPM | TPM | RPD |
|---|---|---|---|
| Alibaba | Unlimited | Unlimited | Unlimited |
| official | Unlimited | Unlimited | Unlimited |
No restriction
Compare qwen3-vl-rerank API pricing, supported endpoints, capabilities and access options on Modelsell.
Create an API key with access to qwen3-vl-rerank, then use the exact model ID and a supported endpoint from the API access section. Request fields depend on the selected endpoint.
Pricing depends on the selected provider group and the model billing unit. The current input, output, request, or media prices are shown on this page before sign-up.
Start with the use cases and prompting guidance on this page, then evaluate the model with representative inputs from your project.
