ante
Ghost in your shell. Ante is a self-contained agent harness with a highly optimized core. It works like Claude Code or Codex, with none of their dependencies or model constraints.
Ante is a lightweight, self-contained AI coding agent built in Rust that operates entirely from the terminal with zero runtime dependencies. The binary compresses to ~15MB and includes an embedded llama.cpp engine for fully offline operation with local GGUF models, eliminating the need for API keys or cloud services. It supports multiple LLM providers (Anthropic, OpenAI, Google Gemini, DeepSeek, and more) while consuming approximately 7× less memory and 9× less CPU than comparable agents. Ante can run in interactive TUI mode, headless for scripting and CI, or as a server daemon for editor integrations.
Key Features
Use Cases
- 01Running AI-assisted coding tasks entirely offline on local hardware without API dependencies
- 02Executing automated code reviews, refactoring, and bug fixes in CI/CD pipelines via headless mode
- 03Building custom agent harnesses with different tool sets and system prompts using named profiles
- 04Integrating AI assistance into editors and IDEs via the server mode JSONL protocol
- 05Deploying resource-efficient coding agents in containerized or memory-constrained environments
- 06Creating chatbot integrations for Slack or Discord using gateway mode
Related Agents
View morehermes-agent
The agent that grows with you
agency-agents
A complete AI agency at your fingertips - From frontend wizards to Reddit community ninjas, from whimsy injectors to reality checkers. Each agent is a specialized expert with personality, processes, and proven deliverables.
openinterpreter
A coding agent for open models like Kimi K3
cline
Autonomous coding agent as an SDK, IDE extension, or CLI assistant.
ante — FAQ
What is Ante?+
Ante is a self-contained AI coding agent built in Rust that runs in your terminal with zero dependencies. It works like Claude Code or GitHub Copilot but without vendor lock-in, supporting any LLM provider or fully offline local models via built-in llama.cpp.
How do I install Ante?+
Run `curl -fsSL https://ante.run/install.sh | bash` to download and install the binary. No additional dependencies or runtime are required. The installer supports custom channels (nightly) and installation directories.
Can Ante run completely offline without API keys?+
Yes. Ante includes a managed llama.cpp inference engine. Point it at any GGUF model file with `ante --offline-model ~/.ante/models/model.gguf -p "your task"` and it runs entirely on your machine without internet access.
Which AI providers and models does Ante support?+
Ante supports 17+ providers including Anthropic (Claude), OpenAI (GPT-5.x), Google Gemini, Grok, DeepSeek, Open Router, and any OpenAI-compatible endpoint. You can add custom providers via `~/.ante/catalog.json` configuration.
Is Ante free to use?+
The Ante binary is free to use during the alpha preview, including for commercial purposes, under the Binary Preview Terms. The SDK, protocol, and documentation are Apache 2.0 licensed. You provide your own API keys or use local models.
How do I integrate Ante with my editor or other tools?+
Run `ante serve` to start Ante in server mode, which communicates over a structured JSONL protocol via stdio or WebSocket. This enables building editor plugins, web UIs, and custom integrations on top of the agent runtime.
How do I install ante?+
Open the source repository on GitHub and follow its README. ante is a agent — MCP Agents Market links you directly to the official repo.
Is ante free?+
ante is an open-source project hosted on GitHub. Check the repository for its license and any usage requirements.