First look at the LEGO Sagrada Família It is the largest LEGO set ever made at 12,060 pieces. Will cost $800.

fal is an official launch partner for @OpenAI's new role-specific plugins in Codex, live today. With fal in Codex, creative teams can pull models, assets, and workflows from fal directly into Codex sessions. Faster from a question or brief to useful output, without leaving the agent.

🚨 Grok Imagine Video 1.5 drops on fal! Turn a prompt or reference frame into polished video in seconds Fluid camera work, coherent scenes and output that holds up on close inspection Built for creators who want premium results without a heavy pipeline

fal Assets is live. Everything you generate on fal lives in one library now: images, video, audio, 3D. Semantically searchable, and instantly reusable as input across Sandbox, Playground, and Workflows. No re-uploading. No filename hunting. Just "the angry one" or "the sunset clip" and you're back in flow.

i made myself a lifeos in 45 minutes which is connected to my whoop band. it's powered by @fal (great way to spend credits is to use openrouter endpoint on fal) i also used @hasaneyldrm 's repo for library of excersizes. what this can do: - pulls recovery, sleep, strain & hrv from whoop - log meals by photo, text or voice (wizper) — describe a meal in natural language and it figures out the kcal macros, web search included - generates training programs from goal days equipment, with 1324 exercise gifs hosted locally - inline edit programs: swap moves, change sets/reps in place - tdee from real whoop kilojoules, not a guess currently only whoop band but can be extended to other fitness trackers as well should i opensource it? would you use it? yet, would you contribute it?

made the sports fan cam trend into @fal workflow. just add your photos and type something like “lakers vs celtics finals game 7” or “wimbledon final center court”

Skills are pre-configured creative workflows you can install in one command. Game avatars, cinematic filmmaking, product renders, UGC and many more. Each skill locks in prompting, styling, structure, and output formats, so agents can generate consistently without manual tuning every run.

Announcing World Model Accelerator, fal's new product: fal.ai/wma We've spent years building the inference layer for generative media. In 2023, we shipped 30 FPS, low-latency, action-controlled diffusion APIs, years before anyone else. Fast forward to today and we've been powering foundational model companies across image, video, real-time speech-to-speech and action-controlled world models. Now we're opening it up. World Model Accelerator is the system behind it all. Our in-house inference engine hits SOTA performance on Hopper and Blackwell for Diffusion Transformer workloads, both causal and bi-directional. Built on fal Serverless, it scales seamlessly from 1 GPU to 1,000 . This is the same platform that's been running real workloads, at scale, for years. A new WebRTC gateway optimizes latency between users and GPUs (routing requests to the closest region), built on the same infra that powers our real-time speech workloads. And direct access to the fal Marketplace puts your model in front of enterprises spending hundreds of millions on generative media, with a real co-selling motion behind it. This is the infrastructure layer for world models.