Getting Started

Model Guides

The Relace API serves our suite of publicly available models for code generation. 

Introduction

Relace

High-speed code generation engine optimized for real-time performance at 2000 tokens/second.

Instant Apply

A specialized code-aware reranking system that eliminates the need for embedding-based retrieval, enabling accurate code edits on large codebases beyond LLM context limits.

Code Reranker

Merge code snippets from an LLM into your existing codebase.

Apply Code

List files in your codebase in order of relevance to a user's query.

Rank Code v1

Assess the relevance of each file in your codebase to a user's query.

Rank Code v2

Embed code snippets into a vector database for semantic search.

Getting Started

Model Guides

Pricing

Introduction

Getting Started

Make an Account

Try the Playground

Model Guides

Instant Apply

Code Reranker

Getting Started

Model Guides

Pricing

​Getting Started

Make an Account

Try the Playground

​Model Guides

Instant Apply

Code Reranker

Getting Started

Model Guides