Model Switcher#

Switch the OpenClaw default model with backup, validation, and automatic rollback.

Workflow#

Step 1: Validate Model Name#

Read ~/.openclaw/openclaw.json. Extract:

• Current model from agents.defaults.model.primary
• Available models from models.providers list

Match user request against available models (exact, alias, or partial match):

• Exact: zai/glm-5 → direct hit
• Alias: "GLM 5" → zai/glm-5; "doubao" → volcengine-plan/doubao-seed-2.0-pro
• Partial: "deepseek" → volcengine-plan/deepseek-v3.2

Step 2: Abort if Not Found#

If no match, stop and report:

• List all available models
• Ask user to pick one

Never guess or execute blindly.

Step 3: Execute Switch#

bash

python3 scripts/model_switch.py <full-model-name> --retry 3

The script handles: backup → modify config → restart gateway → health check → auto-rollback on failure.

Step 4: Expect Session Interruption#

Gateway restart kills the current session. This is normal. The session resumes automatically after Gateway recovers.

Constraints#

1. Only modify agents.defaults.model.primary — never touch the provider list
2. Always validate model exists before switching
3. Backups stored in ~/.openclaw/config_backups/
4. Script has full backup + rollback — config will never be lost