All updates

Google Deepmind is making Gemini Omni 1.2 Flash generally available. The video-generation model now creates clips up to 15 seconds long, can be edited step by step via conversation, and costs just $0.07 per second at 720p, 30 percent less than its predecessor. Users who draft in…

OpenAI is adding a GIF feature to ChatGPT. With the new Images 2.6 model, users can create animated GIFs from text prompts or photos and then adjust speed and individual frames. The model's image consistency is strong enough to keep characters recognizable across frames, making…

Three MiniMax H3 models, three prices. Here's when to use H3 Highspeed, H3 Lite, and H3.

Kling 4.0 makes up to 30 seconds of 1080p video from a prompt, and it can thread one continuous shot through up to 10 keyframes.

For the next two weeks, we’re putting up $25,000 worth of prizes for the best open-source builds made on the Comfy Developer Platform.

Krea’s reference-driven image model — multi-reference inputs, targeted edits that hold the details you name, and fast iteration — lands in ComfyUI as a Partner Node.

5.6B parameters — base and distilled open weights — bounding boxes and multi-reference editing — FLUX 3 Image Klein runs natively in ComfyUI.

Upscale, interpolate, and convert to HDR locally on your RTX GPU, with better EXR support and color management built for VFX pipelines.

Can MiniMax H3 keep up with video playback on one GPU while still producing clips worth watching? ComfyStreamerH3 is my attempt to find out.

1K to 4K output — Minimal, Medium and High thinking — instruction-based editing — Google's Gemini Nano Banana 2.1 lands in ComfyUI as a Partner Node.

The behind-the-scenes video was made to be part of the story.

Describe what you want right inside ComfyUI. It plans, builds and runs workflows.

Package your workflow once, deploy it on the GPU you choose, and call it from any app, tool, or service.

Build once, add models as they launch, and choose where every job runs.

The MiniMax H3 video VAE now encodes up to ~2.2x faster and decodes ~1.4-2.7x faster on Nvidia GPU.

7B parameters, native 2K output, RGBA transparency, and editing across up to 10 input images — Qwen-Image-2.1 runs natively in ComfyUI.

H3 Max generates a 5-second video in under 3 seconds. In our human preference evaluations against twelve leading video models, it ranks first on quality, prompt understanding, and aesthetics. Independent benchmarks from Artificial Analysis and Design Arena rank it first as well…

**OpenAI** announced a proposed Navier–Stokes proof by an internal model "**significantly more capable than GPT-6 Astra**" using **10,000 agents** over **88 hours** plus **17 hours** of formal verification. The effort highlights the emergence of **massive test-time compute…

Comfy experts embedded with your team, building the workflows inside your environment, and training your team to own it!

Two weeks, one rule, and hundreds of entries from nearly 50 countries. Here's who took the four titles, and a look at everyone who made the final ten!

The best open 3D generation models are now part of ComfyUI core: no custom nodes, no dependency hell, no license traps.

Google's updated Omni Flash model brings 4K output, keyframe interpolation, and conversational video editing to your workflow.

Thanks for reading ComfyUI Newsletter!

Announcing H3 Max, a post-trained version of MiniMax H3 developed by fal Research and optimized for maximum speed by fal's inference team. In our human preference evaluations, H3 Max ranks #1 across overall quality, prompt understanding, and aesthetics against leading video…

Alibaba's new video generation model brings 30-second single-pass generation, up to 20 reference assets, and instruction-driven video editing to your workflow.

Two weeks to make something, four ways to win an RTX5090 or RTX5060 Ti, and a special guest judge livestream!

Your agent runs local ComfyUI and builds workflows around your GPU, your models, and your custom nodes. Works from Claude, Cursor, Codex, or any MCP client.

Black Forest Labs' first video model generates audio and video from the same pass, and it takes direction like nothing else we've tried.

At @fal we serve Ideogram V4 at high speed. At 1K resolution, a generation that used to take 2.75 s now renders in 0.44 s, a 6x speedup with no visible quality loss. We got there by pushing the cost of a generated image down on every axis

At @fal , we've achieved 16 times higher throughput on Qwen3.6 for a use-case that required high interactivity per user using DSpark on SGLang. But why did we need it? And how did we achieve it? Intro Text-to-image models like Ideogram V4 can generate stunning,

In a world of slop, creatives need an aesthetics-focused model. Here's how to work with it.

By Emir Lise, Technical Partnerships Manager, AI Infrastructure @fal Today fal announced a strategic partnership with Amazon Web Services (AWS). I interned at AWS in college, working on compute pricing. What surprised me most was how many people were in the room for any given…

If you have never tried a video model before, now is the time.

AI image models have become extremely good at producing something that looks like stone, ceramic, brushed metal, leather, or concrete, but these images are not readily usable in a traditional CGI workflow - they have baked-in specular highlights, occlusion, perspective…

Starting today, HeyGen's models are available on fal. Whether you need talking avatars, automated video generation, or multilingual lip sync, HeyGen brings a focused set of capabilities that complement what's already on the platform. What HeyGen Brings Video Agent API — Give it…

Seedream 5.0 Lite dropped in February 2026 and it's a different beast from what we've been used to. Every single image on this page was generated with the exact prompt shown. This is an independent guide based on our own testing. We just really like

Seedream 5.0 brings multi-step reasoning, example-based editing, and deep domain knowledge to image generation. Here's what you should know.

Recraft V4 generates art-directed images — and actual editable SVGs — with strong composition, accurate text rendering, and what the Recraft team calls "design taste." Four models are available on Replicate now.

FLUX.2 brings professional-grade image generation and editing with unprecedented detail, multi-reference support, and enterprise efficiency.