💸How is bitHuman so cost effective?
Our Essence Model runs entirely on CPU, eliminating expensive GPU costs. Self-hosted Essence is just 1 credit/min (~$0.01/min) — running locally on your own machine means you only pay for the cloud minutes you actually use.
⚙️What's the difference between Essence Model and Expression Model?
Essence Model runs on CPU, supports full-body avatars (humans & animals), and is the most efficient option. Expression Model is GPU-accelerated for high-fidelity facial dynamics, humans only. Both models can be self-hosted (Expression supports either NVIDIA GPU via Docker or Apple Silicon via MPS/CoreML) or used via our cloud API.
🖥️Can I deploy bitHuman agents fully offline?
Yes! Both Essence and Expression support offline deployment — Essence on any CPU, Expression on an NVIDIA GPU (Docker) or Apple Silicon Mac. Run agents entirely offline for privacy and control in sensitive or air-gapped environments.
⚡How fast and easy is deployment?
Deployment is simple and fast. Our SDK is ready-to-go, and the LiveKit plugin makes setup seamless for both cloud and on-premise environments.
🔒Is my data secure and private?
Absolutely. You can run everything offline, and for enterprise clients, we offer custom deployments with full data control and compliance options.
🛠️Can I get a fully custom solution?
Yes! We provide end-to-end agent solutions, including custom model fine-tuning, integrations, and deployment tailored to your business needs.
🔄Can I upgrade or downgrade anytime?
Yes, you can change your plan at any time. Upgrades apply immediately and are prorated. When you downgrade, you keep the credits you've already paid for through the end of your current billing period, then your new plan's allowance applies going forward.
📅What happens to unused credits?
Plan credits reset monthly. Top-up credits roll over to the next month, so you never lose purchased credits.
