A collection of demo projects to help you get started with the TheStage AI platform. Each project is a self-contained example you can clone and run on your containers right away.
pip install thestage
thestage config set --access-token <YOUR_THESTAGE_TOKEN>Get your token at app.thestage.ai -> Profile -> Access Tokens.
Some models require a HuggingFace access token. Set HF_TOKEN as an environment variable in your container configuration.
Get your token at huggingface.co/settings/tokens.
# Run a demo
thestage project run "python main.py"
# Monitor logs
thestage project task logs <task_id>Accelerated AI models produced by TheStage AI ANNA (Automated Neural Networks Accelerator). Each model comes in multiple optimized sizes (S/M/L/XL) trading off speed vs. quality.
| Project | Description | Status |
|---|---|---|
| Flux_schnell_inference | Text-to-image generation with FLUX.1-schnell | Ready |
| TheWhisper_batch_job | Batch audio transcription with TheWhisper | Ready |
| Project | Description | Status |
|---|---|---|
| Qlip_Compile | Model compilation with QLIP | Coming soon |
| Qlip_Compress | Model compression with QLIP | Coming soon |
| Qlip_Serve | Model serving with QLIP | Coming soon |
| Qlip_Skills | Custom skills for QLIP | Coming soon |
All ElasticModels demos require:
- GPU: L40s, RTX 4090/5090, H100, or B200
- Python: 3.10 - 3.12
- CUDA: 12.8+