OpenRouter MCP: Pick, Price, and Test LLMs from Inside Your Coding Agent
Melvin Vivas · X video post · 2026-06-25 · 2:30 · 113 views · Open on X
Topics: AI Agents, Tool Use & MCP, LLM Fundamentals, AI Dev Tools & Productivity · Level: beginner
Summary
This is a product demo of the OpenRouter MCP server. It gives coding agents like Claude Code live data on models: benchmarks, pricing, provider speed, and documentation. That matters because what an agent knows about models comes from training data that is about six months old. The demo filters coding models by price, finds the fastest provider for a chosen model, compares models for design work, and searches OpenRouter's docs for its Python SDK, all without leaving the agent.
Key points
- The problem: coding agents only know the models in their training data, which is about 6 months old, while OpenRouter offers more than 300 models.
- Without the MCP, choosing a model means opening OpenRouter tabs to compare benchmarks, prices, and provider speed by hand.
- Example query: 'What are the top coding models right now, under $2 per million input?' The MCP returns live ranked results.
- You can ask which provider is fastest for a given model. In the demo, Friendli served GLM-5.2 at about 152 tokens/second.
- The MCP can also run test generations with the model you picked, directly inside Claude Code, for example building a landing page.
- A search-docs tool pulls OpenRouter documentation, such as the Python SDK, into the agent's context.
- Setup takes two steps in Claude Code: run
claude mcp add <OpenRouter MCP URL>, then log in through the OpenRouter tool. An OAuth window opens, you approve a key, and you're done. - Ways to save money: filter models by price per million tokens and route to cheaper, capable models instead of overspending on tokens.
Resources mentioned
- OpenRouter MCP · tool · openrouter.ai · free
An MCP server that gives agents live model benchmarks, pricing, provider speed, test inference, and OpenRouter docs search. - OpenRouter · tool · openrouter.ai · free
A single OpenAI-compatible API that routes requests to hundreds of models from many providers, with one bill and model fallbacks.
Also in: Stealth Model 'Space Bunny' Revealed as MiniMax M3.1 Flash (Melvin Vivas on X · notes), NVIDIA Nemotron 3.5 Super VL now on OpenRouter (Melvin Vivas on X · notes), OpenRouter serves image models, including FLUX 3 Image Klein (Melvin Vivas on X · notes), Comparing Opus 5.5, GPT-6 Astra and Grok 4.7 on OpenRouter (Melvin Vivas on X · notes) and 53 more - OpenRouter Documentation · docs · openrouter.ai · free
OpenRouter's docs explaining video generation parameters, the async request flow and the video models route.
Also in: OpenRouter Adds Video Generation: One API for Veo, Seedance, Wan and Sora (Melvin Vivas on X · notes) - OpenRouter Python SDK · tool · openrouter.ai · free
OpenRouter's Python SDK for calling models from your application code. - Claude Code · tool · code.claude.com · paid · recommended by both Bashiri Smith & Melvin Vivas
Build agents and pipelines from the terminal; the guide's main agentic coding tool.
Also in: First impressions: Claude Code in the cloud is good and fast (Melvin Vivas on X · notes), Why Claude Code feels faster: gzipped turns and instant cloud sessions (Melvin Vivas on X · notes), Watch What Fills Your Coding Agent's Context and When to Clear (Melvin Vivas on X · notes), Claude Code's Right Sidebar and Context Breakdown (Melvin Vivas on X · notes) and 124 more - OpenAI Codex · tool · openai.com · paid · recommended by both Bashiri Smith & Melvin Vivas
OpenAI's coding agent. In the diagram it writes code, fixes review findings and drives the build loop. The creator also used it to make this video.
Also in: Developers moving from ChatGPT/Codex to Claude (Melvin Vivas on X · notes), Creator Opinion: Codex Is Still the Best at Computer Use (Melvin Vivas on X · notes), OpenAI Codex Ambassador applications are open (Melvin Vivas on X · notes), GPT-Live-1.1: Multilingual Voice Update for ChatGPT and Codex (Melvin Vivas on X · notes) and 264 more - Cursor · tool · x.com · free · recommended by both Bashiri Smith & Melvin Vivas
AI-native code editor (VS Code with AI built in).
Also in: Finding Cursor Cloud Agents After the Cursor–X Unification (Melvin Vivas on X · notes), Overnight notebook generation with Grok Bot, Cursor cloud and Grok 4.7 (Melvin Vivas on X · notes), Gemini 4 Argon in Cursor: CursorBench Score and Pricing (Melvin Vivas on X · notes), SpaceX AI / Cursor Ambassador Program Waitlist (Melvin Vivas on X · notes) and 135 more - GLM 5.2 · tool · huggingface.co · free
A GLM-family LLM (the transcript says 'GLM-5-2') that the demo ranked as a strong, affordable model for coding and design.
Also in: Qwen3.8-Max on the Frontend Code Arena cost-performance frontier (Melvin Vivas on X · notes), Models That Work With the Hermes Agent for Personal Productivity (Melvin Vivas on X · notes), Open Model GLM 5.2 Helped Mitigate OpenAI-Caused Cyberattack (Melvin Vivas on X · notes), AI News Roundup: Claude Fable 5, Scientist AI, ZCode, NVIDIA RL (Melvin Vivas on X · notes) and 27 more - Friendli · tool · friendli.ai · paid
An inference provider on OpenRouter (the transcript says 'Friendly').
Try this
- Connect the OpenRouter MCP to Claude Code with `claude mcp add <OpenRouter MCP URL>`, then log in through the OpenRouter tool and approve a key in the OAuth window.
- Ask your agent for the top models for your task, filtered by price per million tokens.
- Ask which provider is fastest for the model you chose.
- Use the search-docs tool to bring OpenRouter's Python SDK docs into your agent while you integrate it.
- Build a SaaS landing page with a cheap model you picked through the OpenRouter MCP, then compare the designs from several models.
- Add the OpenRouter Python SDK to a SaaS app so it can route requests to cheaper, capable models and cut token costs.
More in AI Agents, Tool Use & MCP
- Comfy MCP: Controlling ComfyUI Workflows from AI Agents via MCP
- Where to get help with OpenClaw stability issues
- CEO AI Brain: agents answer business questions with charts
- NVIDIA Agent Toolkit: open Nemotron models for domain-specific agents
- Qwen-AgentWorld: a single world model simulating 7 agent environments
- Grok Build /goal: Long-Running Autonomous Tasks with Subagents