Tinker API
  • What is Tinker?
  • Navigating these docs
  • Installation
  • Tinker Support
  • Using the Tinker API
  • Training and Sampling
  • Loss Functions
  • Saving and Loading
  • Downloading Weights
  • Publishing Weights
  • Async and Futures
  • Model Lineup
    • ServiceClient
    • TrainingClient
    • SamplingClient
    • RestClient
    • APIFuture
    • Parameters
    • Exceptions
    • OpenAI
  • The Tinker Cookbook
  • Overview
  • Rendering
  • LoRA Primer
  • Supervised Learning
    • Basic SL
    • SL training loop
    • SL hyperparams
    • Prompt distillation
    • Sweep case study
  • Reinforcement Learning
    • Your first RL run
    • RL envs
    • RL training loop
    • RL hyperparams
    • Sequence Extension
  • Preferences
    • DPO guide
    • RLHF worked example
  • Evaluations
  • Completers
  • Under the Hood
  • Development tips

On This Page

  • Get Tinker Support
Question? Give us feedback → (opens in a new tab)View source on GitHub →
Tinker Support

Support

Get Tinker Support

Our official support email is [email protected] for customer questions and issues. You can also find helpful community discussion on the Tinker Discord (opens in a new tab).

InstallationTraining and Sampling
Tinker API Documentation