“Your private AI-powered visual assistant that goes from spark to stock-ready.”
Designers, marketers, and solopreneurs — this is your fast lane from idea to income.
Modern creators juggle speed, quality, and originality. We built this app to automate the repetitive, spark fresh ideas, and generate consistent visual assets using the latest in AI.
- Instant inspiration: Describe a vibe, theme, or object—get a fully-formed visual prompt.
- Batch generate: Quickly produce multiple visual variations with SDXL-quality output.
- Smart pipeline: From prompt → image → metadata → ready-to-upload files.
- Works offline: Powered by local LLM and ComfyUI, no cloud fees or internet required.
- Graphic Designers
- Art Directors
- Creative Freelancers
- Social Media Managers
- Visual Storytellers
SIA Multi is a no-code desktop app designed to streamline stock image creation by automating prompt generation in bulk from structured CSV inputs.
It is a variant of the Stock Idea Assistant (SIA) with support for multi-line prompt injection, allowing users to batch-generate diverse AI prompts based on CSV task rows — ideal for mockups, themed visuals, and scalable image workflows.
Important: This is a bridge. You will still need to install Ollama & ComfyUI installed on your Machine (SDXL ideally need 8GB VRAM+)
| Features | Free Mode | PRO (Personal) | PRO (Business) |
|---|---|---|---|
| One-Time Cost | USD 0 | USD 29.99 | USD 59.99 |
| Commercial Use | ❌ Not allowed | 🔶 Allowed for personal non-commercial projects | ✅ Allowed for clients, sales, business |
| Use on Multiple Devices | 1 device only | Up to 2 devices | Up to 3 devices |
| Max Prompts/ Images per Batch | 3 | Unlimited | Unlimited |
| Enter "Prompt Theme" Auto Generate Prompts | ✅ | ✅ | ✅ |
| Use Manual Prompts | ✅ | ✅ | ✅ |
| Flatten Prompt Batch | ✅ | ✅ | ✅ |
| CLI ComfyUI Launch | ✅ | ✅ | ✅ |
| Auto Seed Randomization | ✅ | ✅ | ✅ |
| Batch Image Generation (PNG) | ✅ | ✅ | ✅ |
| Batch Image Upscale (RealESRGAN_x4plus.pt) | ✅ | ✅ | ✅ |
| Batch PNG to JPG conversion | ✅ | ✅ | ✅ |
| JPG Metadata Injection | ✅ | ✅ | ✅ |
| Auto Separate PNG/ JPG files | ✅ | ✅ | ✅ |
| One-click FTP Upload | ✅ | ✅ | ✅ |
| Auto Dependency Check | ✅ | ✅ | ✅ |
| Resolution Selection | ✅ | ✅ | ✅ |
| CFG Scale | ✅ | ✅ | ✅ |
| Flexible Folder Mapping | ✅ | ✅ | ✅ |
| Purchase | 🆓 📥Download | 🛒 Purchase | 🛒 Purchase |
✅ All EXE builds are digitally signed with a verified code-signing certificate to ensure authenticity and prevent tampering.
- Ollama
- ComfyUI installed via CLI
- Windows 11 with GPU (Ideally 8GB VRAM and above)
- Tested on both RTX 4060 8GB & RTX 2080TI 12GB
- Image generation speed similar per image (RTX 4060 take a bit longer for comfyUI launch)
- Storage usage:
| Setup Type | Estimate (Min–Max) |
|---|---|
| ComfyUI + SDXL + Upscale | 13.5 – 14.5 GB |
| + With Ollama (Mistral) | 17 – 20+ GB |
| + With Large Ollama models | 22 – 30+ GB |
- GUI and automation bridge only — not the models
- I do not bundle Ollama, Mistral weights, or SDXL models
Q: Can it run on Mac?
A: It is Windows-optimized and not work on Mac.
Q: Do I need internet?
A: Its a standalone exe installation. When installation completed, only need internet for FTP upload e.g. Dreamstime.
Q: What AI model is used?
A: SDXL via ComfyUI + optional Ollama LLM locally + WAS nodes
- Not optimized for mobile workflows (desktop only)
- Even though negative prompts build-in in App, you may still still malform hands
- Includes no prompt limit.
- Self-use.
- Up to 2 devices.
- Includes no prompt limit.
- For clients & business use.
- Up to 3 devices
This free version is licensed for personal, non-commercial use only.
Commercial use requires a Business License.
🔒 See Personal License for full terms.
MLT Stock Image Assistant is an independent tool that assists with AI-based image generation automation. This software does not include or distribute any AI model weights (e.g., Mistral, SDXL), and it assumes the user has properly licensed or downloaded those models independently from their official sources. This tool is not affiliated with Ollama, StabilityAI, Adobe or Dreamstime.