Skip to content

GPTDiff CLI Reference

Core Command Structure

gptdiff "<transformation-prompt>" [FILES...] [OPTIONS]

Key Options

.gitignore and .gptignore

Files matching .gitignore pattern or .gptignore patterns are ignored when no files are specified.

Transformation Control

--apply
AI-powered patch application
Example:
⚠️ Processes files concurrently for performance

gptdiff "Add null safety checks" --apply src/

--call
Generate diff without applying
Example: 

gptdiff "Modernize string formatting" --call

--prepend <file>
Prepend custom instructions from file to system prompt
Example:

gptdiff "Modernize string formatting" --prepend style-guide.txt

--prepend <file_or_url>: Prepend custom instructions from the specified file or URL to the system prompt

--temperature <0-2>
Control transformation creativity
Default: 0.7
Example: 

gptdiff "Refactor legacy API" --temperature 0.3

Model Selection

--model
Choose reasoning engine (default: $GPTDIFF_MODEL or 'deepseek-reasoner')
Options: deepseek-reasoner (structural), gemini-2.0-flash (text)
Example: 

gptdiff "Translate docs to French" --model gemini-2.0-flash

Scope Management

--files
Target specific paths
Example: 

gptdiff "Update config system" config/ utils/config_loader.py

--max_tokens <number>: Set the maximum number of tokens for the API response (default: 30000) --applymodel <model_name>: Specify the model to use for applying the diff (used in smartapply). If not specified, defaults to the model from --model or GPTDIFF_MODEL. --nowarn: Disable the warning and confirmation prompt for large token usage --verbose: Enable verbose output for detailed information during execution

--nobeep
Silence completion alerts
Example: 

gptdiff "Remove deprecated features" --nobeep

Environment Variables

GPTDiff uses the following environment variables: - GPTDIFF_LLM_API_KEY: API key for the LLM service - GPTDIFF_LLM_BASE_URL: Base URL for the LLM API (default: https://nano-gpt.com/api/v1/) - GPTDIFF_MODEL: Default model for generating diffs (default: deepseek-reasoner)

For the smartapply feature, you can set separate variables: - GPTDIFF_SMARTAPPLY_MODEL: Model for smartapply (defaults to GPTDIFF_MODEL if not set) - GPTDIFF_SMARTAPPLY_API_KEY: API key for smartapply (defaults to GPTDIFF_LLM_API_KEY if not set) - GPTDIFF_SMARTAPPLY_BASE_URL: Base URL for smartapply (defaults to GPTDIFF_LLM_BASE_URL if not set)

These allow you to use different models or credentials for generating and applying diffs—perfect for virtual team flexibility!