Update your ComfyUI [Github] to 0.3.72 ( cd comfyui-directory -&gt; terminal -&gt; git pull -&gt; restart ComfyUi)<a rel="ugc" href="https://huggingface.co/unsloth/Mistral-Small-3.2-24B-Instruct-2506-GGUF/tree/main">GGUF text-encoder</a>10GB VRAM + 64GB RAM =&gt; Yes (CLIP device: CPU + <a target="_blank" rel="ugc" href="https://github.com/kijai/ComfyUI-KJNodes">Diffusion Model Loader KJ</a> (Triton &amp; dtype fp8_e4m3fn )<h3 id="recommended-resolutions:">Recommended resolutions:</h3>(Flux2 is flexible and can handle any aspect ratio up to 4MP total)<ul><li>Square: 1024×1024, 2048×2048</li><li>Landscape: 2048×1024, 3072×1536</li><li>Portrait: 1024×2048, 1536×3072</li><li>Ultra-Wide: 4096×1024</li></ul>FLUX.2 [dev] is a 32 billion parameter rectified flow transformer capable of generating, editing and combining images based on text instructions. More info<h3 id="key-features">Key Features</h3><ol><li>State of the art in open text-to-image generation, single-reference editing and multi-reference editing.</li><li>No need for finetuning: character, object and style reference without additional training in one model.</li><li>Trained using guidance distillation, making <code>FLUX.2 [dev]</code> more efficient.</li><li>Open weights to drive new scientific research, and empower artists to develop innovative workflows.</li><li>Generated outputs can be used for personal, scientific, and commercial purposes</li></ol>These models are redistributed here for the sake of convenience.

Flux2-Dev FP8 & GGUF

flux2DevFP8GGUF_textEncoderFP8.safetensors

Text Encoder FP8

GGUF Q5_K_M

GGUF Q2_K

GGUF Q4_K_M

Flux2-Dev VAE

FP8 Mixed

Flux.2 <img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/fb41c6fe-ca73-43ee-82e0-b6ee830afeab/width=525/fb41c6fe-ca73-43ee-82e0-b6ee830afeab.jpeg" />FLUX 2.0 Is Finally Here<a target="_blank" rel="ugc" href="https://flux2.io/flux-2-0-is-finally-here/">Nov 26, 2025</a><a target="_blank" rel="ugc" href="https://flux2.io/">FLUX.2</a> is built for real creative production, not just eye-catching demos. It delivers high-quality visuals with consistent characters and styles across multiple references, follows structured prompts accurately, handles complex text, respects brand guidelines, and manages lighting, layout, and logos with reliability. It can also perform image editing at resolutions up to 4 megapixels while preserving clarity and coherence.<img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/3ea66f47-7c3b-4a27-ae0b-a990edc13b48/width=525/3ea66f47-7c3b-4a27-ae0b-a990edc13b48.jpeg" /><h2 id="black-forest-labs:-open-core">Black Forest Labs: Open Core</h2>BFL Team believe visual intelligence should be developed collaboratively by researchers, creators, and developers everywhere—not concentrated in a few hands. That’s why they pair cutting-edge performance with open research and open innovation. Alongside scalable, customizable production endpoints, we release powerful, transparent, and modular open-weight models for the community.When a team <a target="_blank" rel="ugc" href="https://bfl.ai/blog/24-08-01-bfl">founded Black Forest Labs in 2024</a>, their mission was to make open innovation sustainable, building on their history of delivering some of the world’s most widely used open models. Team combined open systems like FLUX.1 [dev]—now the most popular open-image model worldwide—with professional-grade variants like FLUX.1 Kontext [pro], used by teams from Adobe to Meta. This open-core philosophy fuels experimentation, invites scrutiny, lowers costs, and allows devs to continue sharing open technology from both the Black Forest and the Bay with the world.<h2 id="from-flux.1-to-flux.2">From FLUX.1 to FLUX.2</h2>Where FLUX.1 demonstrated how powerful media models can be as creative tools—delivering precision, efficiency, control, and realism—FLUX.2 shows how frontier-level capability can reshape full production pipelines. By dramatically improving the economics of generation, FLUX.2 is positioned to become a foundational component of modern creative workflows.<img src="https://flux2.io/wp-content/uploads/2025/11/new-910x1024.png" alt="" width="910" height="1024" /><h3 id="what's-new">What’s New</h3><ul><li>Multi-Reference Generation<ul><li>Use up to 10 reference images at once and achieve industry-leading consistency in characters, products, and visual style.</li></ul></li><li>Improved Detail &amp; Photorealism<ul><li>Sharper textures, more precise detailing, and stable, realistic lighting make it ideal for product imagery, visualization, and photography-grade outputs.</li></ul></li><li>Advanced Text Rendering<ul><li>Typography, infographics, memes, and UI mockups now generate with reliably clear, readable fine text suitable for production use.</li></ul></li><li>Stronger Prompt Obedience<ul><li>The model follows complex, multi-section prompts and compositional rules more accurately than before.</li></ul></li><li>Expanded World Knowledge<ul><li>A deeper understanding of real-world context, lighting, physics, and spatial relationships leads to scenes that behave and look the way you’d expect.</li></ul></li><li>Higher Resolution &amp; More Flexible I/O</li></ul>Supports image editing at resolutions up to 4 megapixels, with greater freedom in aspect ratios and input/output formats. Editing: <img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/12bb3f1b-d8d5-49d0-9667-3fe77660d42a/width=525/12bb3f1b-d8d5-49d0-9667-3fe77660d42a.jpeg" /> FLUX.2 [dev]A 32B open-weight model derived from the core FLUX.2 architecture. It is the most capable open-weight model for image generation and editing available today, offering text-to-image and multi-image editing within a single checkpoint. The weights are published on Hugging Face and can be executed locally using our reference inference code. With consumer-grade GPUs (such as GeForce RTX), you can run an optimized fp8 implementation built with NVIDIA and ComfyUI.You can also sample FLUX.2 [dev] through API endpoints on FAL, Replicate, Runware, Verda, TogetherAI, Cloudflare, and DeepInfra.For commercial licensing, visit our website.Source: <a target="_blank" rel="ugc" href="https://flux2.io/flux-2-0-is-finally-here/">https://flux2.io/flux-2-0-is-finally-here/</a>FLUX.2 – VAEA newly engineered variational autoencoder providing a balanced blend of learnability, compression efficiency, and output quality. It underpins all FLUX.2 flow backbones. A detailed technical write-up is available, and the FLUX.2 VAE is released on Hugging Face under the <a target="_blank" rel="ugc" href="https://huggingface.co/black-forest-labs/FLUX.2-dev">Apache 2.0 license</a>.

Files with SHA256: e3467b7d912a234fb929cdf215dc08efdb011810b44bc21081c4234cc75b370e