dspy-finetune-bootstrap

Community

Fine-tune DSPy models with BootstrapFinetune

AuthorOmidZamani
Version1.0.0
Installs0

System Documentation

What problem does it solve?

This Skill distills a DSPy program into fine-tuned model weights for efficient production deployment, reducing inference costs and latency.

Core Features & Use Cases

  • BootstrapFinetune workflow: Prepare a teacher, generate traces, and fine-tune a student model.
  • Production readiness: Output a finetuned program and a saved model path suitable for deployment.
  • Use Case: When you have a large teacher model and want a smaller, fast-serving version without API-only constraints.

Quick Start

Configure a strong teacher LM, instantiate a teacher DSPy module, and run BootstrapFinetune with your trainset and train_kwargs to produce a finetuned model that you can save and load for inference.

Dependency Matrix

Required Modules

None required

Components

Standard package

💻 Claude Code Installation

Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.

Please help me install this Skill:
Name: dspy-finetune-bootstrap
Download link: https://github.com/OmidZamani/dspy-skills/archive/main.zip#dspy-finetune-bootstrap

Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
View Source Repository