Skip to main content
NM

North Model Labs

North Model Labs provides the NORTH Realtime Avatar API, which turns a single reference photo into a photorealistic, lip‑synced video or live WebRTC stream that can be integrated into AI‑driven conversational products. The service runs on a compact model with sub‑second latency, lets developers use their own LLM and text‑to‑speech engines, and charges pay‑as‑you‑go at $7 per hour of rendered video.

Updated 2 months ago

Funding

Funding not disclosed

Funding rounds are not available yet.

Founders

Founder details are not available yet.

Product

Problem

Developers of AI‑driven applications need realistic, low‑latency talking avatars but existing solutions require extensive avatar training, large GPU infrastructure, and costly subscription plans, limiting scalability and flexibility.

Solution

North Model Labs offers the NORTH Realtime Avatar API, a one‑shot service that creates photorealistic, lip‑synced video or live WebRTC streams from a single reference face image. The platform runs on a compact model that can operate on consumer‑grade hardware, delivering sub‑second latency without session caps. Users supply their own large language model, text‑to‑speech engine, and audio, keeping full control over voice, prompts, and data. Pricing is pay‑as‑you‑go at $7 per hour of rendered video, with no credit packs, seat fees, or usage limits. The API includes async job handling, webhook notifications, multi‑viewer tokens, and a hosted React SDK for easy integration into products such as support bots, tutoring systems, or virtual sales agents.

Target Audience

Primary customers are product and engineering teams building AI‑powered conversational interfaces—such as customer support bots, virtual tutors, sales assistants, and interactive companions—that require photorealistic talking avatars.

Features

  • One‑shot avatar generation: single front‑facing photo, no per‑avatar training or fine‑tuning
  • Real‑time lip‑sync via phoneme timing, compatible with any TTS provider and any language
  • Low‑latency WebRTC streaming for live interactive agents, with sub‑second frame updates
  • Unlimited session length and no quality degradation over time
  • Pay‑per‑second billing at $7 / hour, no credit expiration or seat licensing
  • Multi‑viewer tokens enable simultaneous viewing of a live avatar without extra GPU cost
  • Built‑in webhook support and HMAC‑signed payloads for asynchronous job completion
  • Hosted React SDK and comprehensive REST API with examples in Python and Next.js
This profile is AI-generated and may contain inaccuracies.