Setup time: ~10 minutes
Prerequisites
- Warp terminal installed (macOS or Linux)
- Infercom API key
Configuration
Warp’s AI features use sub-agents that can be configured individually.Step 1: Open Warp Settings
- Open Warp
- Press
Cmd+,(macOS) or go to Warp → Settings - Navigate to AI section
Step 2: Configure Custom Model
- Look for Custom Model or BYOK options
- Select OpenAI Compatible as the provider
- Enter the configuration:
- API Base URL:
https://api.infercom.ai/v1 - API Key: Your Infercom API key
- Model:
MiniMax-M2.5
- API Base URL:
Step 3: Configure Sub-Agents
Warp may have multiple AI features that need individual configuration:- Warp AI Chat - Main AI assistant
- Command Suggestions - Inline command completion
- Natural Language to Command - Convert descriptions to commands
Model
UseMiniMax-M2.5 - optimized for agentic coding with 160K context (163,840 tokens) and 75.8% SWE-bench.
Usage
Once configured:AI Chat
PressCtrl+Shift+Space or click the AI button to open Warp AI chat.
Natural Language Commands
Type# followed by a description:
Command Suggestions
As you type, Warp suggests completions based on context.Alternative: Environment Variables
Some Warp AI features may respect OpenAI environment variables:Troubleshooting
Custom Model Not Available
Warp’s BYOK features may require:- Warp Pro subscription
- Latest version of Warp
AI Features Not Using Custom Model
Ensure each sub-agent is configured:- Check all AI-related settings sections
- Some features may have separate model configurations
- Restart Warp after changes
Connection Errors
Verify the API endpoint:Limitations
Next Steps
- Aider - Terminal AI coding assistant
- Goose - Block’s terminal AI assistant
- Choosing a Tool - Compare all options