gpu fine-tune

Manage fine-tuning jobs

Synopsis

Train LoRA/QLoRA adapters on a curated base model: upload a JSONL chat dataset, submit a job, watch its events, and download the resulting adapter when it succeeds. Training runs asynchronously — create returns the job immediately (use --wait-timeout to poll to completion), and get / events track its progress.

gpu fine-tune [command]

Options

  -h, --help   help for fine-tune

Options inherited from parent commands

      --api-base string   API base URL (override with GPUAI_API_BASE env) (default "https://api.gpu.ai/v1")
      --debug             Enable debug logging to stderr
  -o, --output string     Output format: table|json (default table on TTY, json otherwise)

SEE ALSO

• gpu - GPU.ai command-line interface
• gpu fine-tune cancel - Cancel a running fine-tuning job
• gpu fine-tune create - Create a fine-tuning job
• gpu fine-tune download - Download a succeeded job's adapter artifacts
• gpu fine-tune events - List a fine-tuning job's events
• gpu fine-tune get - Show details for one fine-tuning job
• gpu fine-tune list - List your fine-tuning jobs