How AI NPC Platforms Charge - Pricing Models Compared
The sticker price of an AI NPC platform tells you very little. What matters is what is being counted, because that decides what happens to your bill when players actually start talking.
Here is how the main tools charge as of October 2026, and what each model means for a game.
The five pricing models
| Model | Who uses it | What is counted | What grows your bill |
|---|---|---|---|
| Per response | Convai | Each character reply, with monthly quotas and user and concurrency caps | Chatty players, more NPCs, more concurrent users |
| Per minute | ElevenLabs Agents, Charisma.ai | Conversation minutes, with LLM costs on top for ElevenLabs | Long conversations |
| Per token | OpenAI Realtime, Inworld | Audio and text tokens, TTS characters, STT hours | Every second of audio in and out |
| Player pays | Player2 | Each player's own credits and subscription tier | Nothing for you, but players hit limits |
| Monthly plan with credits | Neocortex | Credits included in a monthly plan | Moving up a plan as you grow |
What each model means in practice
Per response. Easy to understand, but quotas and caps matter more than the price. With Convai, concurrency is set per API key, and on a limit of 1 only one user can talk to any NPC at a time. Unused quota does not roll over.
Per minute. Fair for short voice interactions, painful for games where players hang out and chat. ElevenLabs Agents bills per conversation minute and passes LLM costs through on top.
Per token. The most granular and the hardest to predict. OpenAI does not publish a per-minute rate for its realtime models, so you have to measure your own usage to know what a conversation costs.
Player pays. Player2 costs the developer nothing, but every player needs a Player2 account, and when their credits run out your NPCs stop answering.
Monthly plan. Predictable. You know your maximum spend before launch and choose when to upgrade.
How Neocortex charges
- A free plan to build and try characters.
- Paid plans from €9 a month, with commercial use from the Pro plan.
- Monthly credits included in every plan.
- Free testing: talking to your characters in the browser editor does not use credits, on any plan.
- Usage gating tools in the Unity SDK, so you can meter how much each player uses.
- Custom plans for studios that need unlimited usage or self-hosting.
You can also prototype with open models running locally through Ollama in Unity, which costs nothing per message.
Questions to ask any vendor
- What exactly counts as one unit: a reply, a minute, a token, a player?
- Are there caps on concurrent users or monthly active users?
- Does unused usage roll over?
- Is testing during development billed?
- What happens when a quota runs out mid-session?
- Can you run locally or self-host if costs get out of hand?
Start for free
Pricing details were last checked in October 2026 and change often. Always confirm on each vendor's pricing page.
Written by
Sercan Altundas
Date
Oct 17, 2026
Read time
5 min
