gnobitab/InstaFlow

Health v2 · maintenance only

Flags: no_releases

Adoption not part of the score

1409 stars · 47 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

InstaFlow is a one-step text-to-image generation model based on Rectified Flow, enabling ultra-fast Stable Diffusion inference without iterative solvers. It provides pre-trained models, inference code, and demos (Hugging Face, Colab, Gradio) with LoRA and ControlNet compatibility.

Use cases

• generate images from text prompts in a single step
• speed up stable diffusion inference
• run fast text-to-image generation on limited compute
• use one-step diffusion with LoRA models
• use one-step diffusion with ControlNet
• research rectified flow generative models

When to choose

• you need fast, low-latency text-to-image generation
• you want one-step diffusion inference instead of multi-step sampling
• you want to experiment with rectified flow research models

When to avoid

• you need the highest possible image fidelity from multi-step diffusion
• you need a production-grade image generation service out of the box
• you work outside Python/PyTorch environments

Facets

library · maturity active

machine-learning deep-learning image-processing llm-inference artificial-intelligence deep-learning image-processing machine-learning python cross-platform text-to-image stable-diffusion rectified-flow one-step-generation diffusion-models generative-ai iclr-2024 gpu

Member repositories

For agents

markdown · JSON · MCP: product_card(name="gnobitab/InstaFlow")