Solo Commands
Install the uv package manager to acquire solo-server. The following block contains the full installation instructions.Copy
solo --help
╭─ Commands ───────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────╮
│ setup Set up Solo Server environment with interactive prompts and saves configuration to config.json. │
│ robo Robotics operations: motor setup, calibration, teleoperation, data recording, training, and inference │
│ serve Start a model server with the specified model. │
│ status Check running models, system status, and configuration. │
│ list List all downloaded models available in HuggingFace cache and Ollama. │
│ test Test if the Solo server is running correctly. Performs an inference test to verify server functionality. │
│ stop Stops Solo Server services. You can specify a server type with 'ollama', 'vllm', or 'llama.cpp' │
│ Otherwise, all Solo services will be stopped. │
│ download Downloads a Hugging Face model using the huggingface repo id. │
╰──────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯
Start server with SML models
Copy
# Note that you will need Docker for solo serve
solo setup
solo serve --server ollama --model llama3.2:1b