tinker SDK a dedicated training URL backed by one GPU. Your local script controls the training loop. Compute provisions the machine, meters it, and tears it down.
Why use it
Use fine-tuning when a capable base model still misses the behavior your application needs. Supervised fine-tuning teaches from prompt and ideal-response pairs. Reinforcement learning is useful when you can score an answer more easily than you can write the ideal one. The first Compute preset runs LoRA training onQwen/Qwen3-4B-Instruct-2507. It supports cross-entropy for supervised work and importance sampling for RL-shaped loops.
Run a cookbook recipe
Install the current CLI, sign in, and add prepaid credit first:compute tinker run does five things:
- Quotes and provisions a Tinker session.
- Waits for its public training URL to become healthy.
- Injects
TINKER_API_KEY,TINKER_BASE_URL, andTINKER_MODELinto the child process. - Runs the command and forwards its exit status and terminal signals.
- Stops the CLI-owned session and waits for checkpoint harvest.
Run your own script
Your script can build the official client from the injected environment:tinker.ServiceClient() with no arguments also reads the two official environment variables. Do not hard-code the hosted Tinker URL; each Compute session has its own https://tinker-<id>.api.compute.cx base URL.
What you get
The training script receives the normal Tinker futures and results. When the session stops, Compute harvests saved weights to object storage. List the available files with:CLI lifecycle
Choose between one-shot runs, interactive shells, and warm user-owned sessions.
SDK compatibility
Check the supported model and API boundaries, including the SDK environment and losses.
