# Retrieval-augmented generation

This model is distilled from all-MiniLM-L12-v2, achieving nearly 2x faster inference speed while maintaining high accuracy on both CPU and GPU.

ReasonIR-8B is the first retrieval model specifically trained for general reasoning tasks, achieving state-of-the-art retrieval performance on the BRIGHT benchmark and significantly improving performance on MMLU and GPQA benchmarks in RAG applications.

A lightweight text ranking model suitable for reordering search results or documents

A 7B-parameter multilingual text embedding model developed by Alibaba NLP team, specializing in sentence similarity tasks, offering multiple quantization versions

PLLuM-12B-chat is an optimized dialogue version with 12 billion parameters in the Polish large language model family. It is specifically designed for the Polish language and Slavic/Baltic languages, achieving safe and efficient interaction capabilities through instruction fine-tuning and preference learning.

Jina Embeddings V2 Base is an efficient English sentence embedding model, focusing on sentence similarity and feature extraction tasks.

A 3-billion-parameter long-context instruction model fine-tuned based on Granite-3.1-3B-A800M-Base, supporting multilingual tasks

C4AI Command - R is a research version of a high-performance generative model with 35 billion parameters, optimized for various use cases such as inference, summarization, and question-answering.

A large language model with 7B parameters based on Qwen2 architecture, focusing on sentence similarity calculation and text embedding tasks.

An 8-billion-parameter instruction model specifically designed for Thai, with performance comparable to GPT-3.5-turbo, optimized for application scenarios, retrieval-augmented generation, constrained generation, and reasoning tasks

Snowflake Arctic M Long is a sentence-transformers-based sentence embedding model focused on sentence similarity and feature extraction tasks.

A 7-billion-parameter Self-RAG model capable of generating outputs for diverse user queries, adaptively invoking retrieval systems, self-criticizing outputs and retrieved passages, while generating reflection tokens.

SGPT-125M is a sentence transformer model optimized with weighted mean and bitfit techniques, focusing on sentence similarity tasks.