Human direction remains primary.
You set intent, constraints, and standards. The system accelerates the work without taking ownership away from its operator.
Replicant
A new type of AI.
AI Solutions|Designing digital worlds, intuitive user experiences, and powerful automation tools for tomorrow's creators.
Replicant Dossier
Born in the neon sprawl, Mars Ind forges synthetic minds, Replicants engineered to reason, perceive, and act, indistinguishable from human thought but running on silicon fire. Beyond the neural lab, our foundry ships interactive worlds and the tools that build them: original games and developer technology.
- DESIGNATIONAutonomous Cognitive Entity
- ORIGINMars Ind • Neural Lab
- DIVISIONSReplicants • Games • Developer Tools
- CAPABILITYPerception • Reasoning • Action
- STATUSACTIVE: ACCEPTING COMMANDS
We don't just simulate minds, we build the worlds they live in, and the tools that forge them.
ACCESS: COLLABORATION PROTOCOL
COLLABORATION PROTOCOL
Across every division, one principle holds: the human stays in the loop. AI and SI extend judgment, craft, and imagination; they do not hide the work behind a magic button.
Every output is a new input.
Human edits and synthetic transformations move in a deliberate loop until the result earns approval.
Tools amplify understanding.
Code or media produced without domain knowledge is still guesswork, whether it came from a person or a model. Informed operators build reliable systems.
HUMAN Direct + Edit
AI / SI Generate + Transform
HUMAN Judge + Refine
IMAGINGCanvas control + img2img iteration
VOICETimeline craft + synthetic performance
REPLICANTSHuman intent + synthetic cognition
DEVELOPER TOOLSDomain control + intelligent automation
No human degradation. No synthetic mystique. Better work through collaboration.
Games Foundry
Hand-built worlds from the Mars Ind, foundry. Strap in, the asteroid fields don't forgive.
AVAILABLE NOW
Asteroid Navigation
Command the helm of a starship on a perilous voyage to Axios Station. An arcade-style 2D space shooter, dodge treacherous asteroid fields across galaxy after galaxy as the gauntlet tightens. Your reflexes are the only thing between you and the void.
PLATFORM Steam • Windows
MODE Single-player
▶ TRANSMISSION — GAMEPLAY TRAILER
Found a bug or want to share your run? Transmit it straight to the foundry.
MARS IND // GAME COMMERCE NODE
STEAM UPLINK
LIVE PRICING
Developer Tools
Engineered software and developer technology, forged in the same lab as our Replicants; built to drop straight into your pipeline. Select a tool to open its dossier.
A local neural voice studio that runs entirely on your machine; no cloud, no subscription, GPU-accelerated. It pairs two engines in one: Piper TTS drives a wide range of natural voice tones and ages, while XTTS-v2 handles voice cloning and expressive non-word sounds. Assign different speakers to highlighted text for multi-voice scenes, then mix voices with sound effects on a visual audio timeline; the cloning option unites both engines, and that pairing is the heart of the two-engine philosophy.
ENGINE Piper TTS + XTTS-v2 (dual neural)
PLATFORM Windows
AVAILABLE ON itch.io
MODE Desktop Application
🔊 HEAR THE REPLICANT VOICES
Real samples rendered by the engine, tap a Replicant to listen.
18+ Purchase and use of this application is intended for users 18 and older.
System Requirements
Minimum
- OS
- Windows 10 version 22H2 (64-bit)
- Processor
- Intel Core i5-8400 or AMD Ryzen 5 2600
- Memory
- 16 GB RAM
- Graphics
- NVIDIA GeForce RTX 2060 with 6 GB VRAM
- DirectX
- Version 12
- Storage
- 15 GB available space
- Sound
- Windows-compatible audio device
NVIDIA Turing-generation or newer GPU and current NVIDIA drivers required. AMD and Intel GPUs are not supported. A microphone is only required to record new voice-cloning samples.
Recommended
- OS
- Windows 11 (64-bit)
- Processor
- Intel Core i7-10700 or AMD Ryzen 7 3700X
- Memory
- 32 GB RAM
- Graphics
- NVIDIA GeForce RTX 3060 with 12 GB VRAM or better
- DirectX
- Version 12
- Storage
- 20 GB available space
- Sound
- Windows-compatible audio device
An SSD and current NVIDIA drivers are recommended. A microphone is only required to record new voice-cloning samples.
Activation token: Each purchase key activates one machine. To bind your key to a new machine, uninstall the app from the current machine; the uninstaller releases the key. For another simultaneous machine, purchase an additional copy on itch.io to receive another key. All purchases are final.
Multi-Speaker Studio
Type, speak, and direct a whole cast in one script. Highlight any passage, right-click, and assign it a voice; each line is spoken by its own Replicant. Add laughter, hesitation, or spoken physics equations, then render, replay, and export layered conversations and broadcast-ready audio.
Voice Designer
Build a voice from nothing. Powered by Piper, start from an age preset; young child through elderly, then sculpt pitch, pace, warmth, and expressiveness with live sliders across a broad tonal range. Switch on cloning to bridge into XTTS-v2 and turn a sample of a real human voice into a reusable clone.
Voice Cloning Studio
Record a clean sample or import an existing take, inspect the waveform, balance microphone and playback gain, and manage named recordings before cloning. Consent confirmations remain part of the workflow, while full-screen recording controls make it practical to build a reusable voice identity from a focused source sample.
Sound FX
A full ambience engine. Generate rain, storms, and atmospheric effects with fine-grained intensity, density, and brightness controls. Seamless looping crossfades the clip end-to-start so it can play forever with no click, ideal for game ambience.
Audio Timeline
Mix it all together. The Audio tab layers your rendered voices with sound effects on a visual timeline; drag clips to place them, trim the edges, snap them together, and combine everything into one file. Overlapping clips blend, gaps become silence, and the finished mix exports in a click.
Sound Tuning
Dial in the human details. Tune how non-word sounds; laughter, sighs, vocal fillers, and spoken math, are paced and pitched. Set a global default or save bespoke tuning per voice for believable, lifelike delivery everywhere.
Relay Server
Voice as a local service. Spin up an HTTP relay so game engines, scripts, and the Mars relay can request speech on demand. Age-based content safety is enforced, with a simple REST API to list voices and synthesize on the fly.
Feedback welcome, help shape the build.
Off-world image synthesis for indie devs and creators, a fast, flexible, local image generator that delivers reliable results without the hassle. Produce concept art, character sheets, environments, logos and more using multiple DreamShaper models (Lite, Turbo, XL) plus RealVis for photoreal output; pick the tier that fits your rig. A smart img2img workflow creates clean variations, pose changes, and style updates with strong consistency, while a template system saves prompt + seed combinations to reload instantly.
MODELS DreamShaper + RealVis
PLATFORM Windows
AVAILABLE ON itch.io
MODE Desktop Application
18+ Purchase and use of this application is intended for users 18 and older.
System Requirements
Minimum
- OS
- Windows 10 64-bit
- Processor
- Modern 4-core x64 processor
- Memory
- 16 GB RAM
- Graphics
- NVIDIA CUDA-capable GPU with 6 GB VRAM
- DirectX
- Version 12
- Storage
- 45 GB available space
- Internet
- Broadband required for activation and initial engine/model installation
An SSD is highly recommended for model loading and generation performance. Current NVIDIA graphics driver required. AMD and Intel-only GPUs are not supported.
Recommended
- OS
- Windows 11 64-bit
- Processor
- Modern 8-core x64 processor
- Memory
- 32 GB RAM
- Graphics
- NVIDIA RTX GPU with 12 GB VRAM or more
- DirectX
- Version 12
- Storage
- 50 GB available space on an SSD
- Internet
- Broadband recommended for activation, installation, and updates
Current NVIDIA graphics driver required.
Activation token: Each purchase key activates one machine. To bind your key to a new machine, uninstall the app from the current machine; the uninstaller releases the key. For another simultaneous machine, purchase an additional copy on itch.io to receive another key. All purchases are final.
Generation Console
A full studio in one window. Set your seed, steps, CFG, sampler, and screen size, then choose a model tier, DreamShaper Lite, Turbo, or XL, or RealVis for photoreal output. Live Replicant Parameter panels and stackable quality tags let you compose prompts visually and generate straight to a local gallery, no cloud required.
Output Gallery
Every render, organised. Browse your generated frames in a visual gallery, sort them into custom folders by drag-and-drop, and multi-select to move or clean up in bulk. Reload any image straight back into img2img, save it as a template, or export, your whole session stays on-disk and searchable.
Full-Screen Review
Inspect every detail. Double-click any frame to fill the screen, then step through your set with the arrow keys to compare poses and variations at full resolution. A clean, distraction-free viewer for picking the hero shot before you export or send it back into img2img.
Canvas Editor
Edit generated images directly on a layered canvas. Paint and refine masks, add text, transform selections, or use AI Highlight to mark only the region the model may change, then send the result into img2img for further AI-guided editing.
Character Sheet Creation
Generate consistent character references across front, profile, and rear views. Build a production-ready visual sheet for game characters, animation, concept development, and downstream 3D workflows while preserving the core design across angles.
Texture & PBR Materials
Create seamless textures, inspect them on a live 3D material preview, and tune the surface before export. Generate and export a complete PBR material set for use in game engines, 3D scenes, and asset pipelines.
Logo Creation
Develop logos, emblems, badges, and identity marks alongside your other visual assets. Iterate on a direction with the same prompt, seed, variation, and editing tools used throughout the Imaging workflow.
GALLERY
A selection of frames synthesized in Mars Ind - Imaging. Tap any frame to enlarge.
Feedback welcome, help shape the build.
Synthetic signals through the netcode. Mars Ind Relay is the hardline between your AI backend and the outside world, a unified outbound tunnel that binds inference, TTS, image, video, and 3D workloads into a single, disciplined signal stream. It connects to every core you run, establishes a secure tunnel to your gateway, and pushes synthetic signals through the netcode with zero drift.
Relay tracks your inference host automatically, holding a live link even as processes restart, ports shift, or workloads migrate. If the gateway drops, Relay doesn't, it keeps probing the line until the grid comes back online, restoring the tunnel without human intervention. It's built for continuous operation: the kind of reliability you need when your nodes behave like semi-autonomous constructs, and your workloads never sleep. Relay is the signal spine of Mars Ind systems, the conduit that keeps synthetic workloads flowing, even when the grid fractures.
ROLERelay & Outbound Tunnel
PLATFORM Windows
AVAILABLE ON itch.io
System Requirements
Minimum
- OS
- Windows 10 64-bit
- Processor
- Dual-core 2.0 GHz
- Memory
- 4 GB RAM
- Graphics
- Integrated graphics
- Network
- Broadband Internet connection
- Storage
- 2 GB available space
Requires configured gateway and local AI IPs for operation.
Recommended
- OS
- Windows 11 64-bit
- Processor
- Quad-core 2.5 GHz or better
- Memory
- 8 GB RAM
- Graphics
- Integrated graphics or better
- Network
- Broadband Internet connection
- Storage
- 2 GB available space
Requires configured gateway and local AI IPs for operation.
Activation token: Each purchase key activates one machine. To bind your key to a new machine, uninstall the app from the current machine; the uninstaller releases the key. For another simultaneous machine, purchase an additional copy on itch.io to receive another key. All purchases are final.
🖧️ CORE FEATURES
- Outbound tunnel to your gateway for all AI workloads
- Auto-tracking of the inference core host
- API Key Firewall with secure proxy routing
- Auto-reconnect when the gateway goes down
- Automatic stale-session cleanup and RAM recovery
- Unified handling of inference, TTS, image, video & 3D pipelines
- Chaos-resistant throughput for unstable networks
Connection Settings
Point Relay at every core you run, Inference, Neural Voice Synth, Image Forge, Video Synth, and Volumetric 3D Forge, each with its own host, port, and gateway port. Relay auto-tracks the inference host and holds the link even as ports shift or processes restart. No manual SSH required: as long as your gateway has the right IP and port open, the system auto-tunnels straight to it, and power users can point it at any gateway they own.
Tunnel Control & Firewall
Start the tunnel with one command, or arm AUTO Mode so Relay follows your inference core, opening when it's detected, standing down when it's not. The built-in API Key Firewall enforces key-bound access at the relay layer, filtering and proxying every request through a hardened channel so only authorized signals reach your nodes. Automatic memory cleanup clears stale sessions and frees residual RAM, while a live activity log keeps eyes on the line.
feedback directly shapes future versions.