# Requesty

> Unified LLM gateway & secure API for 160+ models including Claude 4, DeepSeek R1, GPT-3Pro.

Requesty routes, secures, and optimizes your LLM traffic — with caching, failover, guardrails, and up to 80% cost savings. Used by 15k+ developers and trusted by teams at Shopify, Appnovation, and Naible.

## Core Pages
- [Homepage](https://www.requesty.ai/): Overview of Requesty's platform and benefits.
- [LLM Routing](https://www.requesty.ai/solution/llm-routing): Route to 160+ models via one API.
- [Smart Routing](https://www.requesty.ai/solution/smart-routing): Auto-selects the best model for each task.
- [Routing Optimizations](https://www.requesty.ai/solution/routing-optimizations): Failover, load balancing, and reliability policies.
- [Integrations](https://www.requesty.ai/solution/integrations): Connect Requesty with your favorite tools.
- [Enterprise Features](https://www.requesty.ai/enterprise): User budgets, SSO, governance, and analytics.
- [Security](https://www.requesty.ai/security): Guardrails, compliance, and incident response.
- [All Available Models](https://www.requesty.ai/solution/llm-routing/models): Complete list of supported models.
- [Blog](https://www.requesty.ai/blog): Deep dives, announcements, and feature releases.

## Core Documentation
- [Quickstart](https://docs.requesty.ai/quickstart): Fast setup with OpenAI-compatible SDK.
- [API Reference](https://docs.requesty.ai/api-reference): All endpoints with examples.
- [Model List](https://app.requesty.ai/model-list): View all available models and usage tiers.
- [Guardrails Overview](https://docs.requesty.ai/features/security): Security & compliance guardrails explained.

## Integrations
- [OpenWebUI](https://docs.requesty.ai/applications/openwebui): Use Requesty in a web-based chat UI.
- [LibreChat](https://docs.requesty.ai/applications/librechat): Bring Requesty to open-source chat UIs.
- [Roo Code](https://docs.requesty.ai/applications/roo-code): Coding assistant with instant model switching.
- [VS Code Extension](https://docs.requesty.ai/applications/VS-code-extension): Switch between GPT-4o and Claude instantly inside VS Code.
- [Cline](https://docs.requesty.ai/applications/cline): Connect Cline AI agent to Requesty.

## Features
- [Smart Routing](https://docs.requesty.ai/features/smart-routing): Automatically route requests to the optimal AI model.
- [Fallback Policies](https://docs.requesty.ai/features/fallback-policies): Configure automatic request re-routing.
- [Load Balancing](https://docs.requesty.ai/features/load-balancing): Distribute your requests across multiple models.
- [Caching & Failover](https://docs.requesty.ai/features/auto-caching): Control automatic caching behavior.
- [Streaming](https://docs.requesty.ai/features/streaming): Receive responses in real-time streams.
- [Structured Outputs](https://docs.requesty.ai/features/structured-outputs): Get consistent JSON responses across different LLMs.
- [Reasoning](https://docs.requesty.ai/features/reasoning): Enable reasoning tokens for enhanced model thinking.
- [SSO & User Spend Limits](https://docs.requesty.ai/features/enterprise): SAML integration + role-based budgets.
- [API Spend Limits](https://docs.requesty.ai/features/api-limits): Set up spend limits for each API key.
- [Bring Your Own Keys](https://docs.requesty.ai/features/bring-your-own-keys): Use your own OpenAI/Anthropic keys.
- [Prompt Library](https://docs.requesty.ai/features/prompt-library): Manage your system and VIbe prompts in one place.
- [Prompt Optimization](https://docs.requesty.ai/features/prompt-optimization): Customize and optimize your system prompts.
- [Request Metadata](https://docs.requesty.ai/features/request-metadata): Add custom metadata to your API calls for powerful analytics.
- [Dedicated Models](https://docs.requesty.ai/features/dedicated-models): Application-specific models for Roo, Cline, coding agents.

## Framework Support
- [LangChain](https://docs.requesty.ai/frameworks/langchain): Using Requesty router with LangChain.
- [Pydantic](https://docs.requesty.ai/frameworks/pydantic): Using Requesty router with Pydantic.
- [Vercel AI SDK](https://docs.requesty.ai/frameworks/vercel-ai-sdk): Official Requesty provider for Vercel AI SDK.

## Examples
- [Prompt Injection Guardrail](https://docs.requesty.ai/features/security#prompt-injection): Real-world redaction example.
- [Fallback Policy Setup](https://docs.requesty.ai/features/fallback-policies#get-started): Create your own model chain.
- [OpenWebUI Setup](https://docs.requesty.ai/applications/openwebui#setup-instructions): Step-by-step walkthrough.
- [API Call w/ Guardrails (Python)](https://docs.requesty.ai/features/auto-caching#python): Example code using auto_cache.

## Optional
- [Blog](https://requesty.ai/blog): Comparisons, announcements & deep dives.
- [Claude 4 Launch Post](https://requesty.ai/blog/claude-4-now-available-on-requesty): Use Claude 4 with fallback and caching.
- [Careers](https://www.requesty.ai/#careers): We're hiring founding engineers.
- [Privacy Policy](https://www.requesty.ai/#privacy): Full compliance and data use policy.
- [Requesty vs OpenRouter](https://requesty.ai/blog/requesty-vs-openrouter): Feature-by-feature comparison.

## Robots & Crawlers
- `User-agent: *`
- `Allow: /llms.txt`
- `Allow: /docs`
- `Allow: /blog`
- `Disallow: /api`
- `Disallow: /internal`

## Attribution / Contact
Maintained by [Requesty](https://requesty.ai).
Contact: [support@requesty.ai](mailto:support@requesty.ai)
Discord: [discord.gg/Td3rwAHgt4](https://discord.gg/Td3rwAHgt4)
API Access: [https://app.requesty.ai/sign-up](https://app.requesty.ai/sign-up)