Local AI Production Pipeline

Direction
to Assets.
One Forge.

InterForge is a local-first AI pipeline for concept artists and game developers. Type a prompt. Get a concept. Refine it to 6 views. Ship the asset. No cloud. No subscription. Your GPU, your files.

01 Prospecting Text → Concept
02 Smelting Concept → 6 Views
03 Forging 6 Views → Mesh ↗
Stage 01 · Prospecting

From prompt
to concept.

Type what you want. SDXL Juggernaut XL generates a batch of concept images tuned to your asset type — character, weapon, prop, tile, environment. Each output comes with a clean RGBA cutout and an SVG silhouette ready for the next stage.

AI Model

SDXL Juggernaut XL v9

Tuned prompt templates per asset type. Style modifiers from painterly to pixel art.

Post-process

RGBA + SVG

rembg background removal. vtracer SVG silhouette. Both locked before Smelt opens.

Asset Types

Built-in Templates

Character, weapon, creature, environment, prop, tile — each with targeted prompt rules.

Styles

Style Modifiers

Painterly, low-poly, pixel art, realistic, stylized — applied per-job via MasterForge.

InterForge · Prospect Stage
// MasterForge config loaded
asset_type character
art_style painterly
lighting studio_neutral
batch_size 4
armored knight, tattered cape, battle-worn,
   full body, front-facing, white bg
── generating ───────────────────────────
step 20/20 ████████████████ done
image_00.png ✓ ready
image_00_rgba.png ✓ masked
image_00.svg ✓ silhouette
→ Smelt unlocked lock image to proceed
Stage 02 · Smelting

One image.
Six views.

Lock a concept and Zero123++ synthesizes six camera angles — front, front-right, right, back, left, front-left — at alternating elevations. Consistent silhouette. Consistent lighting. Each view gets its own RGBA cutout and SVG mask.

Z123

Zero123++ v1.2

Multi-view synthesis with 6 fixed camera poses. Alternating ±20° / −10° elevations for clean 3D coverage.

RGBA

Per-view post-processing

Each of the 6 views: rembg background removal + vtracer SVG silhouette. Sharper masks for reconstruction.

✓×6

Approve per view

Review each view individually. Approve, reject, or re-run a single angle. Only approved views pass to Forge.

In Development
Stage 03 · Forging

Six views.
One mesh.

The approved views feed a visual hull carving pipeline. Silhouette masks carve a 3D volume, photo-consistency refines it, and Open3D Poisson reconstruction produces a watertight mesh. The result is decimated into a 4-level LOD chain and exported as GLB, OBJ, or FBX.

Step 01

Visual Hull

SVG silhouettes from all 6 views carve a 3D volume. The intersection gives the base shape.

Step 02

Photo Refinement

RGB consistency carves concavities the silhouettes can't catch. Front view weighted highest.

Step 03

Poisson + Smooth

Open3D Poisson reconstruction. Taubin smoothing preserves volume — no shrinkage on organics.

Step 04

LOD + Export

4 decimation levels: 100% / 50% / 25% / 10%. Vertex colors projected from the 6 views. GLB out.

GLB — game engine ready
OBJ — universal mesh
FBX — coming soon
LOD×4 — 100 / 50 / 25 / 10%
Vertex Colors — no UV required
Watertight — Poisson surface
Config Engine

MasterForge

The config engine behind every stage. Asset type templates, style modifiers, lighting presets, and reconstruction routes — all selected per-job. Swap a style or a lighting preset and the prompt chain updates automatically.

Asset Types

Character Weapon Creature Environment Prop Tile

Art Styles

Painterly Low Poly Pixel Art Realistic Stylized

Lighting Presets

Studio Neutral Dramatic Soft Rim Light

Reconstruction Routes

Organic Hard Surface 2D Only

MasterForge

Configuration Engine · Per-Job
Organic

Laplacian + Taubin smoothing. Characters, creatures. Preserves volume.

Hard Surface

Sharp edge preservation. Weapons, props, architecture.

2D Only

Disables mesh reconstruction. Sprites, tiles, flat assets.

Negative Prompts Auto per asset type
Elevation (Zero123) ±20° / −10° 6 fixed poses
Smoothing Taubin no volume shrink
LOD Levels 4 100 / 50 / 25 / 10%
Local-First

Your GPU.
Your files.
Your forge.

No cloud, no API keys, no subscription. Every model runs on your hardware. Every output stays on your machine. All job progress streams live over SSE — you see every step in real time.

No Cloud

SDXL and Zero123++ run entirely on your GPU. Nothing leaves your machine.

Live Progress

Every pipeline stage streams real-time SSE events. Step active, step done, image ready — you see it all.

📁

Project Memory

References, sketches, prompts, exports — all saved together per project. Nothing gets lost between sessions.

RequirementSpecNotes
GPU RTX 3070 (8GB VRAM) CUDA 12.1 · target hardware
RAM 16 GB 8 GB minimum
Runtime Python 3.11 + Tauri 2 FastAPI backend · React frontend
3D Viewer Three.js 0.183 OrbitControls · ACES tone mapping
Concept Model SDXL Juggernaut XL v9 via diffusers
Multi-View Model Zero123++ v1.2 6-view synthesis

Tinker Mode — bypass the gates.

The pipeline enforces stage locks by default — Smelt only opens after Prospect, Forge only after Smelt. Tinker Mode drops the gates entirely. Jump to any stage, re-run any step, experiment without committing. Built for rapid iteration when you know what you're doing.

● Active