C

Codesearch ModernBERT Snake

Developed by Shuu12121
A sentence transformer model specifically designed for code search, based on the ModernBERT architecture, supporting 8192 token long sequence processing
Downloads 36
Release Time : 3/26/2025

Model Overview

This model focuses on computing semantic similarity between code snippets and documentation, suitable for code search tasks. Fine-tuned from Shuu12121/CodeModernBERT-Snake, it has the capability to handle ultra-long code sequences.

Model Features

Ultra-Long Sequence Processing
Supports sequences up to 8192 tokens in length, capable of handling extremely long code snippets and documents
Efficient Code Search
Optimized for code search, efficiently computes semantic similarity between code and documentation
Compact Model Size
A small model with only 75 million parameters, yet performs comparably to larger models

Model Capabilities

Code Semantic Embedding Generation
Code-Document Similarity Calculation
Long Code Sequence Processing

Use Cases

Code Search and Retrieval
Code Snippet Search
Search for relevant code snippets based on natural language queries
Achieved a score of 72.12 on the CodeSearchNet benchmark
Document-Code Matching
Automatically match code snippets with their corresponding documentation descriptions
Featured Recommended AI Models
AIbase
Empowering the Future, Your AI Solution Knowledge Base
© 2025AIbase