LLM Orchestration Toolkit

v1.0.0Coding Agentsstable

LLM orchestration toolkit for agent workflows: planner + workers + synthesis, optional router (LLM + learned fallback), supports OpenAI/Anthropic/Ollama/llama.cpp, real scraping with caching, MCP server integration, and a TUI chat UI.

ab-testingai-orchestrationanthropicapi-gatewayartificial-intelligence
Share:
40
Stars
0
Downloads
0
Weekly
0/5

What is LLM Orchestration Toolkit?

LLM Orchestration Toolkit is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to llm orchestration toolkit for agent workflows: planner + workers + synthesis, optional router (llm + learned fallback), supports openai/anthropic/ollama/llama.cpp, real scraping with caching, mcp serv...

LLM orchestration toolkit for agent workflows: planner + workers + synthesis, optional router (LLM + learned fallback), supports OpenAI/Anthropic/Ollama/llama.cpp, real scraping with caching, MCP server integration, and a TUI chat UI.

This server falls under the Coding Agents category on MCPgee, the world's largest MCP server directory with 33,000+ servers.

Features

  • LLM orchestration toolkit for agent workflows: planner + wor

Use Cases

Multi-LLM agent workflows
Planner and worker coordination
Real-time web scraping
llm-use

Maintainer

LicenseMIT
Languagepython
Versionv1.0.0
UpdatedMay 18, 2026
Statushealthy
Maintenanceactive

Works with

ClaudeOpenAIwindowsmacoslinux

Installation

Manual Installation

npx llm-use

Configuration

Configuration Details

Config File

claude_desktop_config.json

Performance

Response Metrics

Response Time< 200ms
ThroughputMedium

Resource Usage

Memory UsageLow
CPU UsageLow

How to Set Up and Use LLM Orchestration Toolkit

llm-use is an LLM orchestration toolkit that implements a planner-plus-workers architecture, routing tasks across multiple AI providers — Anthropic, OpenAI, Ollama, and llama.cpp — in a single coordinated workflow. It includes a learned router with token-vector similarity for automatic model selection, real web scraping with caching, a terminal UI for interactive chat, session history persistence, and an MCP server integration via PolyMCP. Developers building multi-LLM agent pipelines who need provider-agnostic orchestration with cost tracking and offline support will find it a capable foundation.

Prerequisites

  • Python 3.10+ with pip
  • At least one LLM provider: ANTHROPIC_API_KEY for Anthropic, OPENAI_API_KEY for OpenAI, or Ollama installed locally
  • Optional: Playwright for dynamic web scraping, uvicorn for the MCP server mode
  • An MCP-compatible client such as Claude Desktop or Claude Code
1

Clone the repository and install dependencies

Clone the llm-use repository and install the core package along with desired provider libraries.

git clone https://github.com/llm-use/llm-use.git
cd llm-use
pip install -e .
pip install anthropic openai  # cloud providers
pip install polymcp uvicorn    # MCP server support
2

Set provider API keys

Export your API keys as environment variables. At minimum you need one provider configured.

export ANTHROPIC_API_KEY=your_anthropic_key
export OPENAI_API_KEY=your_openai_key
3

Run a basic orchestrated task from the CLI

Test the planner-worker pipeline by running a task with an orchestrator model and a worker model. Cost breakdown is printed at the end.

python3 cli.py exec \
  --orchestrator anthropic:claude-3-7-sonnet-20250219 \
  --worker ollama:llama3.1:8b \
  --task "Summarize the key differences between REST and GraphQL"
4

Launch the TUI chat interface

Start the interactive terminal UI to have a multi-turn conversation routed through the orchestrator-worker pipeline.

python3 cli.py chat \
  --orchestrator anthropic:claude-3-7-sonnet-20250219 \
  --worker ollama:llama3.1:8b
5

Start the MCP server for integration with AI clients

Launch llm-use as an MCP server using PolyMCP and uvicorn so Claude Desktop or other clients can use it as a tool provider.

uvicorn llm_use.mcp_server:app --host 0.0.0.0 --port 8000
6

Add to your MCP client configuration

Register the running llm-use MCP server endpoint in your Claude Desktop or Claude Code config file.

{
  "mcpServers": {
    "llm-use": {
      "url": "http://localhost:8000",
      "env": {
        "ANTHROPIC_API_KEY": "your_anthropic_key",
        "OPENAI_API_KEY": "your_openai_key"
      }
    }
  }
}

LLM Orchestration Toolkit Examples

Client configuration

Connect Claude Desktop to a locally running llm-use MCP server.

{
  "mcpServers": {
    "llm-use": {
      "url": "http://localhost:8000",
      "env": {
        "ANTHROPIC_API_KEY": "your_anthropic_key"
      }
    }
  }
}

Prompts to try

Example tasks you can run via the CLI or after connecting through MCP.

- "Compare 10 products across these URLs and give me a ranked recommendation"
- "Plan and execute a research task on quantum computing advances in 2024 using web scraping"
- "Run a planner-worker pipeline to write a detailed blog post outline on AI safety"
- "Summarize 3 web pages on topic X with caching enabled to avoid re-fetching"

Troubleshooting LLM Orchestration Toolkit

Provider not found or model name error

Use the exact provider prefix format: anthropic:model-name, openai:model-name, or ollama:model-name. Run ollama list to see available local models.

Web scraping fails with dynamic pages

Install Playwright and its browser binaries: pip install playwright && playwright install. Then add --enable-scrape --scrape-backend playwright to your exec command.

Session history not persisting between runs

Session logs are stored in ~/.llm-use/sessions. Ensure the directory is writable. If using Docker or a restricted environment, mount this path as a volume.

Frequently Asked Questions about LLM Orchestration Toolkit

What is LLM Orchestration Toolkit?

LLM Orchestration Toolkit is a Model Context Protocol (MCP) server that llm orchestration toolkit for agent workflows: planner + workers + synthesis, optional router (llm + learned fallback), supports openai/anthropic/ollama/llama.cpp, real scraping with caching, mcp server integration, and a tui chat ui. It connects AI assistants to external tools and data sources through a standardized interface.

How do I install LLM Orchestration Toolkit?

Follow the installation instructions on the LLM Orchestration Toolkit GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.

Which AI clients work with LLM Orchestration Toolkit?

LLM Orchestration Toolkit works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.

Is LLM Orchestration Toolkit free to use?

Yes, LLM Orchestration Toolkit is open source and available under the MIT license. You can use it freely in both personal and commercial projects.

Browse More Coding Agents MCP Servers

Explore all coding agents servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.

Quick Config Preview

{ "mcpServers": { "llm-use": { "command": "npx", "args": ["-y", "llm-use"] } } }

Add this to your claude_desktop_config.json or .cursor/mcp.json

Read the full setup guide →

Ready to use LLM Orchestration Toolkit?

Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.

33,000+ ServersFree & Open SourceStep-by-Step Guides