OpenRouter

2026OpenRouterpaid

Fusion

v1.0
featured
AI
OpenRouter

Fusion is an AI tool developed by OpenRouter that synthesizes the outputs of multiple models to deliver superior performance on complex research tasks. By combining diverse model perspectives, Fusion consistently outperforms individual models, achieving beyond-frontier performance even with budget panels. It is accessible via a single API call, allowing users to leverage the combined strengths of various models seamlessly.

Synthesizes outputs from multiple models
Achieves beyond-frontier performance
Accessible via a single API call
version1.0
regionUnited States
2026OpenRouteropen source

Pony Alpha

v1.0
Agentic AI
LLM

Pony Alpha is a cutting-edge foundation model developed by OpenRouter, designed to excel in coding, agentic workflows, reasoning, and roleplay. It is particularly well-suited for hands-on coding tasks and real-world applications.

Strong performance in coding tasks
Excels in agentic workflows
Advanced reasoning capabilities
pricingFree
version1.0
2026OpenRouterpaid

OpenRouter Response Caching

v1.0
API
Caching

OpenRouter's Response Caching feature caches identical API requests so responses are returned instantly at zero cost — saving time and money on tests, agent retries, and repeated queries.

Zero cost for cached identical requests
Instant response retrieval via X-OpenRouter-Cache header
Available to all OpenRouter users for free
pricingFree
version1.0
2026OpenRouterfree

OpenRouter MCP

Infrastructure
ML

The OpenRouter MCP Server is a component that enables AI coding tools and agents to interact with OpenRouter through the Model Context Protocol (MCP). Rather than expanding a single model's native capabilities, this integration allows your local AI assistant to intelligently select, price, and test the best AI models for a specific task by leveraging OpenRouter's live catalog and benchmark data.

Model-Callable Protocol (MCP) integration
Agent-to-Platform Integration
Live Model Discovery
regionUnited States
hardwareNo specific hardware requirements
2026OpenRouterpaid

Subagent

vBeta
featured
AI
OpenRouter

The Subagent Server Tool is a feature offered by OpenRouter that enables large AI models to delegate specific tasks to smaller, more efficient models during their operation. This delegation allows for optimized performance and resource utilization.

Delegates tasks to smaller, faster models mid-generation
Supports tasks like document summarization and data extraction
Allows worker models to utilize tools like openrouter:web_search
pricingStandard token rates
versionBeta
2026OpenRouterfree

Ori Eval

featured
AI-Evaluation
Model-Selection

Ori Eval is an automated evaluation tool designed to help developers systematically identify the best AI models for their specific applications. It scans your codebase, generates test cases from your prompts, and uses an LLM-as-a-judge to grade performance across metrics like accuracy, latency, and cost.

Automated Evaluation Harness
Context-Aware Testing
LLM-as-a-Judge
pricingFree
regionUnited States
2026OpenRouterfree

Ori Harness

CLI
AI Agents

Ori Harness is a command-line interface tool that allows developers to run existing agent CLIs on OpenRouter without changing their workflow. It provides optimized configurations, unified billing, and organization-wide guardrails for various AI agent harnesses.

Unified billing and activity reporting across all supported agents
OAuth PKCE authentication to eliminate manual API key management
Automatic optimization of agent configurations and environment variables
regionUnited States