skip to content
gigarouter gigarouter
models / reranker · coming soon

Ruri V3 Reranker 310M

cl-nagoya/ruri-v3-reranker-310m

published Apr 2025 · updated Apr 2025

Ruri V3 Reranker 310M is a Japanese general-purpose reranker model built on ModernBERT-Ja, supporting up to 8192 tokens and achieving state-of-the-art performance on Japanese text ranking benchmarks.

est. price
~$0.008
/ 1k docs · estimated, set at launch
API providers
0
downloads / mo
274K
license
apache-2.0

specs

TaskReranking (Text Ranking)
ArchitectureCross-Encoder based on ModernBERT-Ja
Parameters315M
Max Sequence Length8192 tokens
LanguageJapanese
LicenseApache 2.0

about this model

cl-nagoya/ruri-v3-reranker-310m is a Japanese general-purpose reranker model built on ModernBERT-Ja that computes relevance scores between query-document pairs for improved search ranking. It accepts sequences up to 8192 tokens and uses a SentencePiece tokenizer with a 100,000-token vocabulary (up from 32,000 in prior versions). FlashAttention is integrated for efficient inference.

Benchmarks

The model achieves state-of-the-art results on Japanese retrieval benchmarks:
ModelParameters (w/o Emb.)JQaRA nDCG@10JaCWIR MAP@10MIRACL Recall@30
Ruri-v3-reranker-310m315M (236M)86.995.497.3
hotchpotch/japanese-reranker-cross-encoder-xsmall-v1107M (11M)61.493.890.6
hotchpotch/japanese-reranker-cross-encoder-small-v1118M (21M)62.593.992.2
hotchpotch/japanese-rer

best for

FAQ

What is this model best used for?

It is best used for re-ranking search results and improving the quality of retrieved documents in Japanese information retrieval and RAG systems.

How does it compare to other Japanese rerankers?

According to benchmarks, Ruri V3 Reranker 310M achieves higher nDCG@10, MAP@10, and Recall@30 than earlier Japanese reranker models, making it state-of-the-art.

What input format does it expect?

It expects pairs of query and document text. For example, using a CrossEncoder in sentence-transformers with a list of [query, document] pairs.

How can I call it via the gigarouter API?

Use the gigarouter OpenAI-compatible endpoint with your API key, sending a request with a query and a list of documents to be ranked.

What is the license?

The model is released under the Apache License, Version 2.0.

not yet live

We're benchmarking and onboarding Ruri V3 Reranker 310M as a hosted, OpenAI-compatible API. Sign in for free credit and be ready when it lands, or tell us you want it and we'll prioritize it.

related reranker models

compare all →