jina-reranker-v3.5

by Jina AI

jina-reranker-v3.5 is a 0.6B-parameter multilingual listwise document reranker and a drop-in upgrade to jina-reranker-v3. It keeps the same last-but-not-late (LBNL) listwise interface — ranking a query against many candidates in one forward pass — while improving domain robustness, structured-data ranking, multilingual retrieval, and inference efficiency.

API Pricing

Input$0.05 / 1M tokens
Output$0.05 / 1M tokens

Specifications

Context window131,000 tokens
Modalitiestext, image

FAQ

What is jina-reranker-v3.5?

jina-reranker-v3.5 is a 0.6B-parameter multilingual listwise document reranker and a drop-in upgrade to jina-reranker-v3. It keeps the same last-but-not-late (LBNL) listwise interface — ranking a query against many candidates in one forward pass — while improving domain robustness, structured-data ranking, multilingual retrieval, and inference efficiency.

What is the context length of jina-reranker-v3.5?

jina-reranker-v3.5 has a 131,000 token context window.

How much does jina-reranker-v3.5 cost?

On AIHubMix, jina-reranker-v3.5 costs $0.05 per million input tokens and $0.05 per million output tokens.

What modalities does jina-reranker-v3.5 support?

jina-reranker-v3.5 accepts text and image input.

How do I call jina-reranker-v3.5 via API?

jina-reranker-v3.5 is available through the AIHubMix unified API. The API is OpenAI-compatible: point your OpenAI SDK at https://aihubmix.com/v1, use your AIHubMix API key, and set the model name to jina-reranker-v3.5 — no other code changes needed.

Who develops jina-reranker-v3.5?

jina-reranker-v3.5 is developed by Jina AI. AIHubMix aggregates it alongside models from other providers behind one API and one bill.

More from Jina AI

jina-embeddings-v5-text-nano

by Jina AI

A 3.8-billion-parameter general vector model (embedding model) for state-of-the-art…

$0.05/1M in

jina-embeddings-v5-text-small

by Jina AI

A 3.8-billion-parameter general vector (embedding) model providing state-of-the-art…

$0.05/1M in

jina-deepsearch-v1

by Jina AI

DeepSearch combines search, reading, and reasoning capabilities to pursue the best…

$0.05/1M in · $0.05/1M out
1,000,000 tokens context

jina-embeddings-v4

by Jina AI

A general-purpose vector model with 3.8 billion parameters, used for multimodal and…

$0.05/1M in · $0.05/1M out

jina-reranker-v3

by Jina AI

Multimodal multilingual document reranker, 131K context, 0.6B parameters, for visual…

$0.05/1M in · $0.05/1M out
131,000 tokens context

jina-clip-v2

by Jina AI

Multi-modal Embeddings Model, multilingual, 1024-dimensional, 865M parameters.

$0.05/1M in · $0.05/1M out

Use jina-reranker-v3.5 via the AIHubMix unified API — one interface for every major LLM.