Welcome to trtl

Build faster applications with intelligent HTTP response caching that understands the semantic meaning of your requests.

What is trtl?

trtl is a managed serverless semantic cache service built on Cloudflare's edge infrastructure. Unlike traditional caches that require exact URL matches, trtl uses advanced vector embeddings to identify and serve cached responses for semantically similar requests—perfect for AI applications, search systems, and dynamic APIs.

Key Features

• Semantic Similarity Matching - Cache hits based on request meaning, not just exact matches
• Sub-400ms Response Times - Lightning-fast cache lookups at the edge
• 99.999% Accuracy - Reliable semantic matching with configurable confidence thresholds
• Dynamic Infrastructure - Automatic provisioning of storage and vector indexes per cache
• TTL & Size Management - Fine-grained control over cache lifecycle and storage limits
• Multi-Environment Support - Isolated dev, staging, and production environments

Quick Start

Get up and running in minutes with our CLI:

# Install the CLI
npm install -g @trtl/cli

# Authenticate
trtl user login

# Create your first cache
trtl cache create "API Cache" \
  --origin "https://api.example.com" \
  --ttl 3600