Padwan CLI
Padwan CLI is an interactive CLI and TUI for padwan-llm, the unified LLM client library. It provides a terminal interface for querying multiple LLM providers — OpenAI, Gemini, Mistral, and Grok — through a single tool.
Playground project
This is a playground for experimenting with padwan-llm features (streaming, agents, MCP, thinking tokens, batch jobs) — not a production-grade tool. Expect rough edges and breaking changes.
Features
- One-shot queries — send a prompt and get a response, with optional streaming
- Interactive chat — multi-turn conversations with persistent session history
- Batch processing — submit, poll, and export Gemini batch jobs
- Dual interface — works as both a traditional CLI and a rich TUI (via piou)
- Multi-provider — switch between providers with a
-mflag
Quick example
# Try it without installing
uvx padwan-cli "Explain monads in one sentence" -m gpt-4o-mini
# One-shot query
padwan-cli "Explain monads in one sentence" -m gpt-4o-mini
# Stream the response
padwan-cli "Write a haiku about Rust" -m gpt-4o-mini --stream
# Interactive chat
padwan-cli chat send "Hello!" -m gpt-4o-mini
# List available models
padwan-cli models
# CLI mode (no TUI) — same commands, plain stdout
python -m padwan_cli "Explain monads" -m gpt-4o-mini
The padwan-cli script launches the TUI; python -m padwan_cli runs the same commands in plain CLI mode for scripting and piping. See Getting Started for details.
Commands overview
| Command | Description |
|---|---|
| (default) | One-shot LLM query |
models |
List available models across providers |
info |
Show model count per provider |
chat send |
Start an interactive conversation |
chat clear |
Clear conversation history |
batch create |
Create a Gemini batch job |
batch status |
Check batch job status |
batch list |
List recent batch jobs |
batch poll |
Poll a batch job until completion |
batch cancel |
Cancel a batch job |
batch retry |
Retry failed requests from a batch |
batch export |
Export batch results to a file |