ElBruno.Text2Image.Cli 1.7.0

t2i — AI Text-to-Image CLI

Cross-platform CLI for AI text-to-image generation. This is the Lite edition — cloud providers only, no local inference.

Quick Install

dotnet tool install --global ElBruno.Text2Image.Cli --source https://api.nuget.org/v3/index.json

Verify installation:

t2i --version

Quick Start

First-time setup wizard (interactive):

t2i config

Generate an image with your configured provider:

t2i "a robot painting a landscape"

The image is saved as output.png (or a timestamped file). Use --provider to switch between FLUX.2, MAI-Image-2.5/2.5-Flash, MAI-Image-2.6/2.6-Flash (Preview), GPT-Image-1.5, or GPT-Image-2.5.

Providers (Lite Edition)

This Lite edition includes cloud providers only:

Why Lite?

• Small package: ~30 MB (vs ~200 MB with local ONNX Runtime providers)
• No ONNX Runtime native libraries: Keeps installation fast and lightweight
• Cloud-first: Ideal for CI/CD, containers, and environments without local GPU

Want Local Inference?

For local CPU/GPU inference with Stable Diffusion models, see the upcoming ElBruno.Text2Image.Cli.Full package (planned for v0.2.0).

Usage

# Generate with default provider
t2i "a futuristic cityscape"

# FLUX.2 Pro (photorealistic)
t2i --provider foundry-flux2 "a robot painting a landscape"

# FLUX.2 Flex (text-heavy design)
t2i --provider foundry-flux2 --set foundry-flux2.model=FLUX.2-flex "a business card with modern design"

# MAI-Image-2.5 and 2.5-Flash
t2i --provider foundry-mai25 "a photograph of a red fox in an autumn forest"
t2i --provider foundry-mai25-flash "a quick concept sketch of a city skyline"

# MAI-Image-2.6 and 2.6-Flash (Preview)
t2i --provider foundry-mai26 "a product photo" --auto-aspect-ratio --web-grounding
t2i --provider foundry-mai26-flash "a concept sketch"

# GPT-Image-1.5 (Azure OpenAI)
t2i --provider foundry-gpt-image-1p5 "an impressionist painting of a garden"

# GPT-Image-2.5 variants (Azure OpenAI)
t2i --provider foundry-gpt-image-25-sunburst "a sci-fi space station in orbit"
t2i --provider foundry-gpt-image-25-flare "a futuristic city at sunset"

# Custom dimensions
t2i "abstract art" --width 1024 --height 1024

# Custom output path
t2i "a sunset" --out sunset.png

# Image-to-image with reference images (FLUX.2, GPT-Image, MAI-Image-2.5)
t2i "make this photo look like a watercolor" --provider foundry-flux2 --image photo.jpg
t2i "replace the sky with a sunset" --provider foundry-gpt-image-2 --image room.png --mask sky-mask.png

Configuration

Set up cloud providers:

# Interactive setup
t2i config                       # Wizard for all providers
t2i config set foundry-flux2     # Setup FLUX.2
t2i config set foundry-mai25     # Setup MAI-Image-2.5
t2i config set foundry-gpt-image-1p5  # Setup GPT-Image-1.5
t2i config set foundry-gpt-image-25-sunburst  # Setup GPT-Image-2.5-Sunburst
t2i config set foundry-gpt-image-25-flare     # Setup GPT-Image-2.5-Flare

# Bulk edit: apply a shared apiKey / endpoint to ALL cloud providers at once
t2i config set-all apiKey <your-api-key>
t2i config set-all endpoint https://<your-resource>.services.ai.azure.com

# Microsoft Foundry providers (FLUX.2, MAI-Image-2.5)
export T2I_FOUNDRY_FLUX2_ENDPOINT="https://your-resource.services.ai.azure.com"
export T2I_FOUNDRY_FLUX2_APIKEY="your-api-key"

# Azure OpenAI providers (GPT-Image-1.5, GPT-Image-2.5)
export T2I_FOUNDRY_GPT_IMAGE_1P5_ENDPOINT="https://your-resource.openai.azure.com/"
export T2I_FOUNDRY_GPT_IMAGE_1P5_APIKEY="your-api-key"
export T2I_FOUNDRY_GPT_IMAGE_25_SUNBURST_ENDPOINT="https://your-resource.openai.azure.com/"
export T2I_FOUNDRY_GPT_IMAGE_25_SUNBURST_APIKEY="your-api-key"
export T2I_FOUNDRY_GPT_IMAGE_25_FLARE_ENDPOINT="https://your-resource.openai.azure.com/"
export T2I_FOUNDRY_GPT_IMAGE_25_FLARE_APIKEY="your-api-key"

List available providers:

t2i providers

Run diagnostics:

t2i doctor

Initialize AI agent skill file:

t2i init

This creates .github/skills/t2i/SKILL.md and .claude/skills/t2i/SKILL.md in your repository so AI coding agents (GitHub Copilot, Claude Code) automatically know how to use the t2i CLI.

Documentation

Full documentation: docs/cli-tool.md

Repository: github.com/elbruno/ElBruno.Text2Image

License

MIT — see LICENSE