Metadata-Version: 2.4
Name: alakazam-mira-mini
Version: 0.1.0
Summary: Play MIRA — a neural world model of car soccer — locally on your own GPU (CUDA, Apple Silicon).
License: MIT
Requires-Python: >=3.10
Requires-Dist: einops>=0.8
Requires-Dist: fastapi>=0.110
Requires-Dist: huggingface-hub>=0.23
Requires-Dist: numpy
Requires-Dist: omegaconf>=2.3
Requires-Dist: opencv-python-headless>=4.9
Requires-Dist: pillow
Requires-Dist: pydantic>=2.7
Requires-Dist: termcolor
Requires-Dist: torch>=2.6
Requires-Dist: torchmetrics
Requires-Dist: tqdm
Requires-Dist: uvicorn[standard]>=0.27
Requires-Dist: websockets>=12
Description-Content-Type: text/markdown

# alakazam-mira-mini

Play **MIRA** — a neural world model of car soccer, every frame generated live by the model —
locally on your own GPU. No cloud, no account; after the first weight download everything runs
on your machine (NVIDIA CUDA, or Apple Silicon via MPS).

```
pip install alakazam-mira-mini
mira-mini play
```

Our from-scratch reproduction of the MIRA recipe (General Intuition × Kyutai), distilled
(fewer diffusion steps, smaller student, compact decoder) until it runs on consumer hardware.
