Jina AI: 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 que…

jina/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 a single forward pass — while improving domain robustness, structured-data ranking, multilingual retrieval, and inference efficiency. Built on Qwen3-0.6B with 28 layers using hybrid 3L2G attention (sliding-window layers, window 1024, interleaved with global-attention layers and a pinned terminal global layer required by LBNL) and a lightweight MLP projector (1024→512→512), it supports up to 131K tokens of context and ranks documents jointly via causal self-attention and cosine scoring. It reaches 63.20 nDCG@10 on BEIR — above the 4B Qwen3-Reranker at roughly 7× fewer parameters — while running 1.22×–1.56× faster than v3 across short-to-long contexts.

Specifications & capabilities

Input
text
Output
rerank
Context window · tokens
131072
Published model ID
jina-reranker-v3.5
Capabilities
rerank

Published prices

1–1 / 1

Provider specifications & access

1–1 / 1

jina ↗

source model id
jina-ai/jina-reranker-v3.5
hosting
jina-hosted
platforms
  • Jina AI
context length
131072
architecture
input_modalities
  • text
output_modalities
  • rerank
capabilities
  • rerank

Source specification observations

1–1 / 1

jina ↗

field
context_length
value
131072
normalized value
131072
scope
source-model
source id
jina
confidence
1

Original sources

1–6 / 6

Jina AI · Model page ↗

Hugging Face · jinaai/jina-reranker-v3.5 ↗

arXiv · Paper · 2509.25085 ↗

arXiv · Paper · 2607.18152 ↗

Jina AI · Source snapshot ↗

Jina AI · Catalog ↗

Source links

1–2 / 2

Explore more models

Models

Loading
EVIDENCE / COMPARISON

Benchmarks

Explore every benchmark

Loading

These charts contain model-linked results. The charts above also include unlinked source profiles. Compare scores within each evaluation. Model quality, usage, transparency and system performance measure different things.

Install TensorTrace