For weeks now a secret image editing model has been making waves on LMArena. The consensus speculation has been Google was behind the so called "nano-banana" and today this speculation has been confirmed. nano-banana is Google's Gemini 2.5 Flash Image editing model and we are

Mirelo and fal are teaming up to bring complete experiences to AI video creation. With Mirelo SFX seamlessly integrated into fal’s platform, every video gains perfectly synced, production ready sound effects with no extra prompts or effort required. Together, we are making it effortless for creators and enterprises to

Recently Alibaba released Qwen Image a powerful and versatile image model. However this was just a preview of their ultimate plan: Qwen Image Edit, a SOTA open weights image editing model. Text Qwen Image Edit excels at text editing, beyond what we have seen from previous image editing models. For

In creative production, the difference between good and great often comes down to two things: having the right tool for the job, and getting it in place fast. That’s exactly what fal’s extensive model library delivers - instant access to hundreds of high-quality, ready-to-run AI models, from proven

We are proud to announce fal's inference and training endpoints Alibaba's Qwen Image model. It is has been an incredible month for generative media. We've been blessed not one, but two unbelievable high quality image models Wan 2.2 14b and Qwen Image. I

We are excited to announce Marey Realism v1.5 by Moonvalley, now available on fal. Marey Realism v1.5 is trained exclusively on fully licensed, high-definition content. This breakthrough technology empowers filmmakers and visual creators to produce professional-grade video content with unprecedented control and commercial freedom. What is Marey Realism

Alibaba released the Wan 2.2 family of video models recently, but they buried the lede: Wan 2.2 is probably the best open-source image model in the world today. We are excited to announce we have added both Wan 2.2. text to image inference and training on fal.

Consistent Characters from a Single Image We are excited to announce Ideogram Character, now available on fal. Ideogram Character enables developers and creators to generate consistent, photorealistic characters from just a single reference image. This capability opens new possibilities for consistent storytelling, creative projects, and character-driven content. What is Ideogram

Recently through a partnership with Black Forest Labs, Krea has a released a open weights fine-tuned version of FLUX.1, FLUX.1 Krea [dev]. fal has shipped API endpoints for inference and training at blazing fal speeds. Exploring the Flux.1 Krea [dev] When FLUX.1 was released last year

Today, we are excited to share that we’ve raised a $125M Series C led by Meritech, with new participation from Salesforce Ventures, Shopify Ventures, and the Google AI Futures Fund. Our existing partners Bessemer Venture Partners, Kindred Ventures, a16z, Notable Capital, First Round Capital, Unusual Ventures, and Village Global,

The architectural revolution that makes cinematic-quality video generation accessible at scale The launch of WAN 2.2 marks a watershed moment for AI video generation. While WAN 2.1 showed us what was possible, WAN 2.2 delivers what's practical—professional video synthesis that runs on hardware you

The infrastructure revolution that makes enterprise-grade video synthesis accessible at scale WAN 2.2 has just launched, bringing revolutionary AI video generation capabilities that developers can integrate today. This comprehensive guide covers everything you need to know about WAN 2.2's API, from its groundbreaking Mixture-of-Experts architecture to

We’re excited to welcome two powerful Bria models to fal: Bria 3.2 Text-to-Image and Bria Video Background Removal. These models set a new standard for creative teams who demand both quality and responsibility from AI. Bria 3.2 Text to Image: Stunning, Compliant, and Ready to Build * Visuals

We are excited to announce a major update to LTXV (LTX-Video) by Lightricks, now the first gen-Video model capable of long-form generation, with high controllability out-ranking all open source models in lightning-fast speed. This breakthrough introduces a new class of creative storytelling tools for developers and creators. LTXV 0.9.

Open-weights, Fast Inference and LoRA Support Following last month's successful launch of FLUX.1 Kontext [pro] and [max] models, we're excited to announce the release of BFL's FLUX.1 Kontext [dev] with open weights. This new version delivers exceptional performance with immediate fast inference

We’re excited to welcome VEED to the fal platform — bringing three powerful AI models designed to make avatar-driven video creation easier, faster, and more scalable than ever. Whether you’re building content for social media, marketing campaigns, training videos, or internal communications, VEED’s models let you create polished

Veo 3 is the latest state-of-the-art video generation model from Google DeepMind, and it's now first available as an API through fal. You can start creating with Veo3 today at https://fal.ai/models/fal-ai/veo3. This revolutionary model brings three game-changing capabilities to AI video generation: * Improved

Today, we’re excited to announce our partnership with Resemble AI, bringing their powerful suite of real-time, high-fidelity text-to-speech and voice cloning models to the fal platform. Introducing Resemble AI on fal Resemble AI sets a new benchmark in TTS by combining zero-shot voice cloning, expressive speech synthesis, and real-time

We are thrilled to announce a new frontier image editing is now available at fal. FLUX.1 Kontext just joined our platform, and this isn't your average image model. This multimodal flow model understands both text AND images, letting you: * Preserve character consistency across multiple scenes (seriously, check

We're thrilled to announce that PixVerse v4.5 is now live and available on fal. Try it out on the Pixverse v4.5 playground! This latest release delivers significant improvements to video quality, motion responsiveness, and creative control—empowering creators, developers, and storytellers with enhanced tools to push

fal is excited to announce our partnership with Tavus, bringing the Hummingbird-0 AI Lip Sync Model to our platform. Experience the state-of-the-art, zero-shot lip sync capabilities of Hummingbird today in the fal model gallery. Introducing Hummingbird-0 Lip Sync Developed by Tavus as a research artifact from their Phoenix-3 model, Hummingbird

We're excited to announce that FASHN's Try-On v1.5 model is now available on the fal platform. To ensure a smooth transition, we plan to run both v1.5 and the current v1.0 side-by-side for a month before retiring v1.0, giving you time to

We're thrilled to unveil a major update to the Kling family, featuring Kling 2.0 image-to-video and text-to-video. This update is designed to empower creators with an unparalleled level of artistic and editing control. These powerful models offer a rich palette of effects and features to transform your

The Finegrain Eraser is a breakthrough model that goes beyond simply deleting objects from images. It also removes their visual footprints - like shadows, reflections, or lighting - and fills in the missing pixels with hyper-realistic background reconstruction. All you need is a prompt. The Finegrain Eraser comes in different

fal is thrilled to announce our partnership with Sync Labs to bring you Lipsync 2.0, the world’s most natural video-to-video lipsyncing model. To experience Lipsync 2.0, visit the model gallery on fal today. Introducing Lipsync 2.0 Introducing Lipsync 2.0, a zero-shot lipsyncing model that preserves