OpenHuman FAQ — Common Problems and How to Fix Them

This FAQ covers the most common OpenHuman issues and their solutions.

Installation Issues

App won't open after installation

Solution: Check system requirements (Windows 10+ / macOS 12+ / Linux kernel 5.x+). Try re-downloading the installer. On macOS, check Security & Privacy settings to allow the app.

Command line install fails

Solution: Ensure curl is installed. Try running with sudo: curl -fsSL https://get.openhuman.cn | sudo bash. Check that your system has all required dependencies (Node.js 24+ is needed for source builds).

Connection Errors

OAuth integration fails

Solution: Disable ad-blockers and VPNs during the authorization flow. Make sure you're logged into the correct account. Try a different browser.

Auto-Fetch not syncing

Solution: Check that the service integration is enabled (Settings → Integrations). Verify the sync_interval_minutes setting in config.toml is reasonable. Check your internet connection.

Model Problems

Model not responding

Solution: Verify your API key is correct and has available credits. Check that the base_url endpoint is correct. For Ollama, ensure the service is running (ollama serve).

Slow responses

Solution: Try a smaller model (gpt-4o-mini instead of gpt-4o, or a 3B local model instead of 8B). Reduce the context window in TokenJuice settings. Check your internet speed for cloud APIs.

High API costs

Solution: Switch to a cheaper provider (DeepSeek costs 90% less than OpenAI). Enable TokenJuice compression. Use model routing to send simple queries to cheaper models.

Memory Tree Issues

Memory Tree not growing

Solution: Ensure Memory Tree is enabled in config: [memory_tree] enabled = true. Be more specific in your conversations — detailed discussions produce richer memories.

Can't find old memories

Solution: Make sure auto-prune isn't too aggressive. Check max_depth setting. Try asking more specific questions with context.

General

Is OpenHuman really free?

Yes! OpenHuman itself is free and open source. You only pay for third-party LLM API calls if you use cloud models. With Ollama, you pay nothing.

How do I update OpenHuman?

Check for updates in Settings → About. Or re-run the installation command. On macOS, the app auto-updates via the built-in updater.

Can I use OpenHuman without internet?

Yes, if you use local models (Ollama). All features except cloud API calls work offline.