Deploy
Pick a model, see how many nodes can run it right now and the cheapest live hourly rate, then deploy it on one of them.
LLMs
3 templatesOllama
Run open-weight models locally behind an OpenAI-compatible API. Pull a model and serve it on the node.
no nodes available for this model right now
Docker image
llama.cpp server
Minimal GGUF inference server. Lightweight footprint, well suited to small nodes and quantised models.
no nodes available for this model right now
Docker image
vLLM
High-throughput inference server with an OpenAI-compatible endpoint. Best fit for sustained multi-user serving.
no nodes available for this model right now
Docker image
Image generation
2 templatesComfyUI
Node-graph interface for image diffusion workflows. Good for batch generation and pipeline experiments.
no nodes available for this model right now
Docker image
Stable Diffusion WebUI
The familiar AUTOMATIC1111 web interface for Stable Diffusion, with img2img, inpainting and extensions.
no nodes available for this model right now
Docker image
Speech & audio
1 templatesWhisper API
Speech-to-text HTTP service wrapping Whisper, with transcription and translation endpoints.
no nodes available for this model right now
Docker image
Developer & rendering
2 templatesBlender headless
Blender in background mode for scripted renders, EEVEE/Cycles jobs and batch scene processing.
no nodes available for this model right now
Docker image
PyTorch + Jupyter
PyTorch with CUDA, paired with a Jupyter notebook server for interactive training and evaluation.
no nodes available for this model right now
Docker image