Task
Text Ranking
R
cl-nagoya/ruri-v3-reranker-310m
Text Ranking·cl-nagoya· 274.0K· 14
sentence-transformers
apache-2.0
Text Ranking
315.2M params
dataset:cl-nagoya/ruri-v3-dataset-rerankerarxiv:2409.07737base_model:cl-nagoya/ruri-v3-pt-310mbase_model:finetune:cl-nagoya/ruri-v3-pt-310mlicense:apache-2.0
Ruri-v3 Reranker is a general-purpose Japanese reranker model built on top of ModernBERT-Ja. Ruri v3 offers several key technical advantages: - State-of-the-art performance with good robustness for vaious domains - Supports sequence lengths up to 8192 tokens - Expanded vocabulary of 100K tokens, compared to 32K in v1 and v2 - Integrated FlashAttention, following ModernBERT's architecture - Tokeniz
# pull & run locally
pip install mlforge-sdk && mlforge pull cl-nagoya/ruri-v3-reranker-310m
pip install mlforge-sdk && mlforge pull cl-nagoya/ruri-v3-reranker-310m
Model details
Provider
cl-nagoya
Framework
sentence-transformers
Parameters
315.2M
Size
1.5 GB
License
apache-2.0
Downloads
274.0K
Likes
14
Paper
arXiv:2409.07737
Updated
2025-04-18
About cl-nagoya/ruri-v3-reranker-310m
Ruri-v3 Reranker is a general-purpose Japanese reranker model built on top of ModernBERT-Ja. Ruri v3 offers several key technical advantages: - State-of-the-art performance with good robustness for vaious domains - Supports sequence lengths up to 8192 tokens - Expanded vocabulary of 100K tokens, compared to 32K in v1 and v2 - Integrated FlashAttention, following ModernBERT's architecture - Tokenizer based solely on SentencePiece
Related Text Ranking
M
cross-encoder/ms-marco-MiniLM-L6-v2
85.1M 278
🤗 HF
M
cross-encoder/ms-marco-MiniLM-L4-v2
3.8M 16
🤗 HF
M
cross-encoder/ms-marco-MiniLM-L12-v2
2.5M 106
🤗 HF
G
Alibaba-NLP/gte-reranker-modernbert-base
2.1M 95
🤗 HF
Q
Qwen/Qwen3-Reranker-0.6B
2.0M 367
🤗 HF
J
jinaai/jina-reranker-v2-base-multilingual
1.8M 351
🤗 HF