Pre-seed round open. Backing the persistence layer for game worlds?
The persistence layer
for game worlds.
Identity, relationships, and state that carry across every session, so what players do keeps mattering. AI Characters are the first product built on it, with more world systems built on the same foundation.
Works with your engine
Game backends persist progress. They don’t persist the world.
A save file keeps flags. It does not keep what happened between people. With MistScale, one moment keeps shaping the world days later.
Save file only
Day 1
The session ends. The save keeps gold and position.
Day 3
The guard greets the player like a stranger.
Day 9
The blacksmith is as friendly as ever.
Day 14
Nothing that happened between them carried forward.
With MistScale
Day 1
Entity stateThe blacksmith remembers the threat.
Day 3
Information flowThe guard has heard about it, unverified.
Day 9
RelationshipsThe blacksmith's trust toward the player has dropped.
Day 14
AI CharactersWhen the player returns, both characters bring it up in their own way.
Your world at the center. The platform around it.
MistScale keeps the entities of your world, their state, and everything between them persistent. What it stores, it can also pass on, so systems build on the same foundation instead of each keeping its own copy of the truth.
MistScale
Your world
Players, characters, and everything between them
Identity
Every player and every NPC has one persistent identity.
Entity state
Memory, relationship state, and identity state for each entity.
Relationships
Trust, patience, and mood per player. Directional trust between NPCs.
Information flow
Facts pass to the characters that are trusted enough. One hop.
AI Characters
NPCs that read and write the state of the entities around them.
Also part of the platform: scene context (location, time of day, weather) sent from your game in real time.
Built on the platform
AI Characters are the first product built on the platform.
The clearest proof of persistent state is a character that knows who just walked in. Memory, mood, relationships, and lore are one consumer of the same foundation.
Persistent memory
Each character keeps what happened with each player, across sessions.
Mood that moves slowly
A fast per-message emotion over a mood baseline that shifts one step at a time.
Per-player relationships
The same character can trust one player and be wary of another.
Grounded in your lore
Grounding checks specific claims against memory, lore, and the character brief.
No characters in this project yet
Interactive preview with sample data.
Build your first characterModel research
We are training our own model.
MistScale AI runs on a cognition layer we built: retrieval, grounding, relationship state, mood. A general model sits at the center of it today. Varen is ours: a language model fine-tuned for game cognition, so the character work lives in the model rather than in the prompt around it.
Working name
Varen
A language model fine-tuned for game cognition and in-world characters.
- Type
- Fine-tuned LLM
- Built for
- Game cognition and NPCs
In development
Built for the loop, not for chat
The job is not open conversation. It is reading persistent state, grounding an answer in it, holding character, and deciding what was worth remembering. That is what it is tuned on.
Grounding as an objective
Today a separate pipeline checks specific claims against memory, lore, and the character brief. We are training that behaviour in rather than bolting it on.
State as a native input
Per-player memory, trust, and mood are already structured on our side. A model trained on that shape does not need it flattened into a prompt first.
We are not publishing weights, evaluations, or a date. When there is something worth measuring, we will show the working.
One fact, many readers.
The state is the product. One moment in your game becomes a fact that more than one system reads. This is one example of state that persists and moves, not the whole platform.
One fact
A player burns the village mill.
It reaches MistScale through conversation. Sharing follows per-character rules and trust thresholds.
- 01
Entity state
The player boasts about it to the tavern keeper, who stores it as a memory about that player.
- 02
Relationships
The tavern keeper's trust and mood toward the player shift over time.
- 03
Information flow
The tavern keeper is set to Open and trusts the miller at 6 out of 10, so the miller receives the fact as unverified hearsay.
- 04
AI Characters
When the player next talks to the miller, he can bring up what he heard.
For developers
Connect your game in minutes.
Engine SDKs for Unity, Unreal, and Godot, plus Web, REST, and WebSocket. State lives on the server, so there is nothing to save or restore on the device.
1using Mistscale.SDK;2using UnityEngine;34public class TavernKeeper : MonoBehaviour5{6 [SerializeField] private MistscaleNPC mira; // drag in the Inspector78 void Start()9 {10 mira.OnNPCResponseReceived += reply => dialogueUI.Show(reply);11 }1213 // call this from your chat input or interaction key14 public void TalkTo(string message)15 {16 mira.SetSpatialContext("The Hollow Lantern", weather: "rain");17 mira.SendChat(message);18 }19}
Built for worlds that last.
Persistent RPGs
Characters remember each player across a long campaign, and word of what they did gets around.
Survival and sandbox worlds
Settlers and traders keep track of who helped them and who did not.
Social simulations
Relationships between characters decide who hears what, and from whom.
Co-op worlds
Every player in the party builds a separate relationship with the same character.
Player-driven worlds
Players shape a world where relationships, knowledge, and consequences carry forward.
Persistent multiplayer worlds
Give every player a world that remembers what happened, across sessions and across characters.
