Model Selection

One-time Model Override

ask -m llama3.2:3b "summarize this"

Persist Default Model

Both flags are equivalent:

ask -ms
ask -sm

Direct persistence without the picker:

ask -ms llama3.2:3b

Stored in:

• ~/.config/askllm/config.json (unless ASK_CONFIG_FILE is set)

Power Model Flow (OpenAI)

ask -pc gpt-5.4
ask -p "analyze this architecture and provide migration phases"

• -pc sets a dedicated power model
• -p uses the configured power model for current invocation

Missing API Key Handling

If a selected provider key is missing, the CLI can prompt you to paste a key and retry. You can also preconfigure all provider/env keys with:

ask --env-setup