Documentation

Myrlin Docs

Generate 3D assets for Hytale and Minecraft from text or images. Use the web app or integrate with our API.

What is Myrlin?

Myrlin turns text descriptions and reference images into game-ready 3D assets. It outputs native .blockymodel files for Hytale and .glb files for Minecraft, Blockbench, and other engines.

You can use Myrlin through the web app at forge.myrlin.io, or programmatically through the REST API.

Fabricator

Interactive asset creation with live 3D preview

Asset Library

Manage, download, and export your creations

REST API

Generate assets programmatically at scale

Using the Fabricator

Text to 3D

Describe what you want to create. Myrlin generates a concept image first, then converts it into a textured 3D model. You can iterate on the concept before committing to 3D generation.

Enter a prompt describing your asset
Review the concept image. Iterate if needed.
Approve and generate the 3D model
Choose your voxel resolution and convert to BlockyModel
Download in your preferred format

Image to 3D

Upload a reference image instead of typing a prompt. Myrlin uses it as the basis for 3D generation. Works best with clean, single-object images on a simple background.

Retexturing

Apply new textures to an existing 3D model. Useful for creating color variants or upgrading texture quality on completed assets.

Asset Library

All generated assets are saved to your library automatically. From there you can:

Preview models in an interactive 3D viewer
Download as .blockymodel, .glb, .bbmodel, .obj, .stl, or .vox
Generate resolution variants of the same model
Rename and organize your creations

Your assets are private by default. Featured assets in the Gallery are selected by admins from community submissions.

Export Formats

.blockymodel

Native Hytale format. Import directly into Hytale Model Maker or use in mods.

.glb

Standard 3D format. Works with Blender, Unity, Unreal, and most 3D tools.

.bbmodel

Blockbench project file. Open and edit in Blockbench with full box structure.

.obj

Wavefront OBJ with MTL. Universal mesh format for legacy tools.

.vox

MagicaVoxel format for voxel editing and rendering.

.stl

3D printing format. Export your game assets for physical printing.

Texture Atlas (.png)

Auto-generated texture map paired with your model file.

Pricing

Myrlin uses a token system. Each generation step costs tokens. New accounts receive free tokens to get started.

Action	Tokens
Generate concept image	1
Iterate on concept	1
Generate 3D model	4
Retexture	3
API generation (full pipeline)	7

Tokens can be purchased individually or through a subscription plan for volume discounts and monthly allocations.

API Reference

The Myrlin API lets you generate 3D assets programmatically. Authenticate with an API key, start a generation, and poll for results.

Authentication

Pass your API key in the X-API-Key header. Keys start with myrl_ and are created in Settings. API access requires an active subscription.

X-API-Key: myrl_your_key_here

Base URL

https://api.myrlin.io/v1

Endpoints

Method	Path	Description	Cost
POST	/v1/generate	Start a generation	7 tokens
GET	/v1/generations/{id}	Poll status and download URLs	Free
GET	/v1/balance	Check token balance	Free
GET	/v1/usage	Usage history	Free
GET	/v1/usage/stats	Aggregated usage stats	Free

POST /v1/generate

Start a new generation. Returns immediately with a generation ID. Poll GET /v1/generations/{id} for progress and results.

Field	Type	Required	Description
prompt	string	Yes	What to generate (3-500 chars)
workflow	string	No	text_to_3d (default) or image_to_3d
style	string	No	hytale (default) or minecraft
reference_image_base64	string	For image_to_3d	Base64 encoded reference image
voxel_size	number	No	0.1 (32px), 0.067 (48px), 0.05 (64px)
api_version	int	No	Set to 2 to enable v2 features (target_size). Default: 1
target_size	object	No (v2)	Target dimensions in Hytale units. {"x": 65} for uniform scale, {"x": 65, "y": 12, "z": 3} for per-axis

GET /v1/generations/{id} Response

When status is complete, download URLs are populated. Poll every 10 seconds.

{
  "id": "abf628cc-20e1-46ea-989d-e021f31769ce",
  "status": "complete",
  "progress": 100,
  "prompt": "A steampunk pistol with brass gears",
  "glb_url": "https://...",
  "blockymodel_url": "https://...",
  "texture_url": "https://...",
  "thumbnail_url": "https://...",
  "tokens_charged": 7
}

Status values: pending, generating, complete, failed. Download URLs expire after 24 hours. Re-poll to get fresh URLs.

Examples

Python

import requests, time

API_KEY = "myrl_your_key_here"
BASE = "https://api.myrlin.io/v1"
HEADERS = {"X-API-Key": API_KEY, "Content-Type": "application/json"}

# Start generation
gen = requests.post(f"{BASE}/generate", headers=HEADERS, json={
    "prompt": "A crystal sword with ice enchantment",
    "style": "hytale"
}).json()

# Poll until complete
while gen["status"] != "complete":
    time.sleep(10)
    gen = requests.get(f"{BASE}/generations/{gen['id']}", headers=HEADERS).json()
    print(f"{gen['progress']}%")

# Download
for key in ["glb_url", "blockymodel_url", "texture_url"]:
    if gen.get(key):
        data = requests.get(gen[key]).content
        with open(f"model.{key.split('_')[0]}", "wb") as f:
            f.write(data)

JavaScript

const API_KEY = "myrl_your_key_here";
const BASE = "https://api.myrlin.io/v1";
const headers = { "X-API-Key": API_KEY, "Content-Type": "application/json" };

// Start generation
const gen = await fetch(`${BASE}/generate`, {
  method: "POST", headers,
  body: JSON.stringify({ prompt: "A crystal sword", style: "hytale" }),
}).then(r => r.json());

// Poll until complete
let status = gen;
while (status.status !== "complete") {
  await new Promise(r => setTimeout(r, 10000));
  status = await fetch(`${BASE}/generations/${gen.id}`, { headers })
    .then(r => r.json());
}

console.log("GLB:", status.glb_url);
console.log("BlockyModel:", status.blockymodel_url);

cURL

# Start generation
curl -X POST "https://api.myrlin.io/v1/generate" \
  -H "X-API-Key: myrl_your_key_here" \
  -H "Content-Type: application/json" \
  -d '{"prompt": "A crystal sword", "style": "hytale"}'

# Poll status
curl "https://api.myrlin.io/v1/generations/GENERATION_ID" \
  -H "X-API-Key: myrl_your_key_here"

# Check balance
curl "https://api.myrlin.io/v1/balance" \
  -H "X-API-Key: myrl_your_key_here"

V2: Target Dimensions (Python)

Set api_version to 2 and provide target_size to control output model dimensions. The model is rescaled after voxelization to match your target.

import requests, time

API_KEY = "myrl_your_key_here"
BASE = "https://api.myrlin.io/v1"
HEADERS = {"X-API-Key": API_KEY, "Content-Type": "application/json"}

# Generate a sword scaled to 65 units along X axis
gen = requests.post(f"{BASE}/generate", headers=HEADERS, json={
    "prompt": "A Persian scimitar sword for Hytale",
    "style": "hytale",
    "api_version": 2,
    "target_size": {"x": 65}  # Uniform scale: longest axis = 65 units
}).json()

# Or specify all three axes independently
gen = requests.post(f"{BASE}/generate", headers=HEADERS, json={
    "prompt": "A tower shield for Hytale",
    "style": "hytale",
    "api_version": 2,
    "target_size": {"x": 8, "y": 40, "z": 30}  # Per-axis dimensions
}).json()

# Poll and download as usual
while gen["status"] != "complete":
    time.sleep(10)
    gen = requests.get(f"{BASE}/generations/{gen['id']}", headers=HEADERS).json()

Error Handling

Status	Meaning
401	Invalid or missing API key
402	Insufficient token balance
403	Access denied (wrong owner or no subscription)
404	Generation not found
429	Rate limit exceeded (check Retry-After header)

If generation fails after tokens are charged, they are automatically refunded.

Start Building

Try the Fabricator or create an API key to generate assets programmatically.