| --- |
| title: README |
| emoji: 🚀 |
| colorFrom: blue |
| colorTo: indigo |
| sdk: static |
| pinned: false |
| --- |
| |
| # Runcrate |
|
|
| **GPUs and inference for your agents.** [runcrate.ai](https://www.runcrate.ai/?utm_source=huggingface&utm_medium=directory&utm_campaign=gpu_pricing) |
|
|
| - **Inference:** open models (DeepSeek, Qwen, Llama, Kimi, FLUX, Whisper) behind one OpenAI-compatible API at `https://api.runcrate.ai/v1`, priced per token, image, second or minute depending on the model. [Models and prices](https://www.runcrate.ai/inference/pricing?utm_source=huggingface&utm_medium=directory&utm_campaign=gpu_pricing) |
| - **GPU marketplace:** rent H100, H200, A100, L40S, RTX 4090 and more with SSH, priced per GPU-hour and billed by the second on most GPUs, by the minute on a few. [GPU pricing](https://www.runcrate.ai/pricing?utm_source=huggingface&utm_medium=directory&utm_campaign=gpu_pricing) |
| - **Built for agents:** a remote MCP server at `https://mcp.runcrate.ai/mcp` (OAuth sign-in, no key) lets Claude Code, Cursor or Codex launch a GPU, run commands on it and shut it down. [MCP setup](https://www.runcrate.ai/docs/mcp/setup?utm_source=huggingface&utm_medium=directory&utm_campaign=mcp) |
| - **Tools:** CLI (`brew install aeonmindai/tap/runcrate`), Python (`pip install runcrate-sdk`) and TypeScript (`npm i @runcrate/sdk`) SDKs. |
|
|
| Try the models in the [inference playground](https://huggingface.co/spaces/runcrate/inference-playground), or run the [Colab quickstart](https://colab.research.google.com/drive/1r5gpncUfDCjHwHOd53nypbblCDSLvips) (OpenAI client, LangChain, LlamaIndex, DSPy). |
|
|
| ```python |
| from openai import OpenAI |
| |
| client = OpenAI(base_url="https://api.runcrate.ai/v1", api_key="RUNCRATE_API_KEY") |
| print(client.chat.completions.create( |
| model="deepseek-ai/DeepSeek-V3.2", |
| messages=[{"role": "user", "content": "Hello"}], |
| ).choices[0].message.content) |
| ``` |
|
|
| Contact: hello@runcrate.ai |
|
|
