coding agents
Source control with out-of-the-box codebase retrieval, fast utility SLMs, and task-specific agents you can run on any repo.
Sign up for free
Everything you need for autonomous codegen
Repos
Source control designed for agents with light weight push/pull operations and no rate limits.
Code Retrieval
Best in class semantic search that scales to large codebases out-of-the-box.
Fast Apply
Universal code merging model that applies file edits at 10,000 tok/s.
State of the art SLMs as tools for coding agents
Small, fast models trained in-house to outperform frontier LLMs on utility tasks. Equip your agent with tools to apply file edits at 10,000 tok/s, and search an entire codebase in less than 2s.
Built for production
Low-latency inference, SOC 2 compliant, encryption in transit and at rest. Deploy on our cloud, your VPC, or on-premise.
Everything you need
Specialized Models
Purpose-built SLMs for embed, rerank, and apply operations.
Fast Retrieval
Two-stage semantic search finds code in under 2 seconds.
Smart Merging
10,000 tokens per second with conflict resolution.
Low Latency
Optimized inference stack for real-time responses.
Simple Integration
Drop-in API works with your existing CI/CD pipeline.
Built for Reliability
SOC 2 compliant with enterprise-grade security.
Specialized Models
Purpose-built SLMs for embed, rerank, and apply operations.
Fast Retrieval
Two-stage semantic search finds code in under 2 seconds.
Smart Merging
10,000 tokens per second with conflict resolution.
Low Latency
Optimized inference stack for real-time responses.
Simple Integration
Drop-in API works with your existing CI/CD pipeline.
Built for Reliability
SOC 2 compliant with enterprise-grade security.
Trusted by the trailblazers
Relace has been a critical tool for us to create custom, design-focused AI models and allowed us to fine tune on our own data and continuously create better and better models.
Just wanted to say that Relace's fast rewriting model has been a big boon to our product. It's made edits much more reliable with hardly any visible downsides. A whole class of bugs is gone for us now.
Do you guys ever sleep?
This is definetely a lot faster!
Relace has been a critical tool for us to create custom, design-focused AI models and allowed us to fine tune on our own data and continuously create better and better models.
Just wanted to say that Relace's fast rewriting model has been a big boon to our product. It's made edits much more reliable with hardly any visible downsides. A whole class of bugs is gone for us now.
Do you guys ever sleep?
This is definetely a lot faster!
Frequently asked questions
Our Apply model processes 10,000 tokens per second. Search completes in under 2 seconds for codebases of any size.
We offer hosted API, self-hosted deployments, and VPC-isolated options for enterprise customers.
Yes, we're SOC 2 Type II certified with encryption in transit and at rest.
We offer a free tier for testing and development. Enterprise pricing is based on usage and deployment type. Contact us for details.
Our infrastructure supports custom model deployments for enterprise customers. Reach out to discuss your requirements.
Get Started in Minutes
Try out our playground, and start with our free tier to test Relace models in your application.
Sign up for free