Mxbai Rerank Large V2
1.5B-parameter RL-trained reranking model achieving state-of-the-art accuracy across 100+ languages with 8K context, outperforming Cohere and Voyage.
About model
Mxbai Rerank Large V2 is a powerful reranker model offering state-of-the-art performance and strong efficiency, with multilingual support for 100+ languages. It is suitable for users seeking a simple end-to-end retrieval solution.
Quickstart guides
Model card
🌟 Features
- state-of-the-art performance and strong efficiency
- multilingual support (100+ languages, outstanding English and Chinese performance)
- code support
- long-context support
Benchmark Results
*Latency measured on A100 GPU
Training Details
The models were trained using a three-step process:
- GRPO (Guided Reinforcement Prompt Optimization)
- Contrastive Learning
- Preference Learning
- Model providerMixedbread AI
- TypeRerank
- Main use casesRerank
- DeploymentServerlessOn-Demand Dedicated
- Parameters1.5B
- Context length8192
- Input price
$0.10 / 1M tokens
- Input modalitiesText
- Output modalitiesText
- ReleasedMarch 12, 2025
- External link
- CategoryRerank