I

Inf Retriever V1 1.5b

Developed by infly
INF-Retriever-v1-1.5B is a dense retrieval model based on large language models developed by INF TECH, optimized and fine-tuned for Chinese-English data retrieval tasks.
Downloads 19.59k
Release Time : 2/8/2025

Model Overview

This model is a lightweight version based on gte-Qwen2-1.5B-instruct, specifically optimized and fine-tuned for Chinese-English data retrieval tasks, achieving excellent performance in the AIR-Bench benchmark.

Model Features

Chinese-English Retrieval Optimization
The model is specifically optimized for Chinese-English retrieval scenarios, significantly improving the accuracy and efficiency of various retrieval tasks.
Top-Tier Performance
Achieved leading scores in the AIR-Bench benchmark, making it an ideal choice for cross-domain heterogeneous information retrieval.
Large Context Support
Supports input lengths of up to 32,768 tokens, making it suitable for long-document retrieval tasks.

Model Capabilities

Text Embedding
Information Retrieval
Sentence Similarity Calculation
Cross-Language Retrieval

Use Cases

Information Retrieval
Web Search Queries
Given a web search query, retrieve relevant answer paragraphs.
Achieved excellent performance in the AIR-Bench benchmark.
Cross-Domain Retrieval
Supports retrieval tasks across multiple domains such as Wikipedia, web pages, healthcare, law, academia, news, and finance.
Performed well in cross-domain retrieval across 13 languages.
Featured Recommended AI Models
AIbase
Empowering the Future, Your AI Solution Knowledge Base
© 2025AIbase