flux2

A fast AI Video Generator for the GPU Poor. Supports Minimax H3, LTX-2.3/2.5, Wan 2.1/2.2, Qwen Image 1/2, Krea 2, Hunyuan Video, LTX Video and Flux.

A GitOps workflow example for multi-env deployments with Flux, Kustomize and Helm.

Weave GitOps is transitioning to a community driven project! It provides insights into your application deployments, and makes continuous delivery with GitOps easier to adopt and scale across your teams.

Flux Subsystem for Argo

Manage multi-tenant clusters with Flux

The open .char format for consistent characters. Face, body, wardrobe & voice in one file, consistent across every model.

Create high-fidelity Codex companions from 2+ photos with local open-weight neural editing and no OpenAI API key.

GitOps Visual Studio Code Extension

GitOps principles to define kubernetes cluster state via code. (Read-only Mirror)

Community maintained Helm charts for Flux

ComfyUI custom node for NVIDIA PiD pixel diffusion decoding and upscale workflows

This repository contains infrastructure applications and add-ons installed in Kubernetes via FluxCD v2.

First PuLID implementation for FLUX.2 — Consistent face identity in ComfyUI

Split FLUX.2 and LTX 2.3 across two GPUs (LAN or same-machine) — NVENC compresses activations live on the wire. Icarus (ComfyUI node) + Daedalus (back-half server).

Identify kubernetes resources which are not managed by GitOps

WeeLLM runs large diffusion models with as little as 4 GB of VRAM, without any quantization. It dynamically determines how many layers can fit within the available VRAM and streams the text encoder and transformer layers to the GPU layer by layer, enabling inference on hardware with limited VRAM. It supports both safetensors and GGUF models.

Source repo of Building SaaS applications on Amazon EKS using GitOps workshop

My home Kubernetes cluster

flux-2 API (flux2 / flux 2): 1MP $0.024; 2MP $0.036; 3MP $0.048. Model id, prompts settings, curl and Python examples over an OpenAI-compatible gateway, $1 minimum top-up.

From dataset to trained LoRA to finished image or video. One browser workspace for your models, datasets, training, and generation. LoRA Pilot supports 50+ training model groups, with ComfyUI, Kohya, AI Toolkit, and Diffusion Pipe included. Run on your own GPU or RunPod, and connect AI agents through MCP.