it's FINALLY here! introducing flowith. human thoughts aren't linear, they're multithreaded. flowith brings you the future of human-ai interaction. flowith is our 12-month journey to redefine human-ai interaction. current chat uis are just the beginning. we're crafting the platform for gpt-5 and future agi. we're committed to radical innovation. experience the next evolution in ai interfaces. flowith features: • intuitive canvas: multi-thread convos & output comparison • oracle: ai agent system for step-by-step, complex problem-solving • 30 tools integrated, 1000 coming soon • ai-driven workflows: auto-design, real-time tweaks • top ai models: gpt-4o, claude 3.5, llama 3.1 & more meet oracle: our cutting-edge ai agent system breaks down complex tasks with finesse. it auto-generates subtasks, taps into tools like twitter and youtube search, webpage generator - all expandable. plus, you can tweak the workflow. each step builds on previous results, diving deeper with every interaction. flowith 1.0 is live! we believe in building, not hype. experience the future now at flowith.io and share your thoughts. just launched on product hunt - your upvotes mean the world to us! producthunt.com/posts/flowit… retweet or upvote for a free month of advanced plan. dm screenshot for access. let's reshape ai interaction together.

tried opus 4.8. fast mode is finally priced right. dynamic workflows look so crazy i have to go deep on this. codex on pause, no regrets.

codex got noticeably nerfed past few days. ran several tasks for 20 hours, none finished. switched to claude code, done in 30 min. something's off @thsottiaux

if you're an ai founder still building agent harness, you're already behind. the game is harness for harness. honestly probably harness for harness for harness. the meta-layer is the moat.

Imagine if flowith existed in 1995

seedance 2.0 canvas cowork = a 1m-view travel vlog factory. 20 chinese street foods. no passport. no uploaded images. just tell your agent what to make. it writes every prompt and batch-generates all 20 videos on one canvas. Install: npx skills add flowith-ai/canvas-cowork like, retweet, and reply. we’re giving away 20 free memberships to those who try the workflow. here’s what it produced 👇

we get a new agent framework every 12 hours but almost no one is building evals that keep up. a lot of this is vibe coded. how do we even know what's good?

we said Flowith Canvas is “native to the agent mind.” here's what that actually looks like. introducing the canvas cowork skill: give one command to your agents (claude code, codex, etc.) → watch them collaborate across the canvas to batch-generate dozens of images and videos. 50x faster for visual output. 0 manual routing. get started: npx skills add flowith-ai/canvas-cowork or github ⬇️

two years ago: the first ai infinite canvas for humans. now: a creative space where any agent can show up and build. claude code, openclaw, neo — one skill, infinite collaboration.

3.1 Lightning. intelligent. tiny. efficient.

This is insane!!! Really like the voice mode!!! @flowith @DerekNee

daytime: work with the best agents side by side — claude code three we built ourselves (browser, computer use, video editing). nighttime: hand your machine over, earn while you sleep.

new OS still in beta. already 100x better than our last version. not another openclaw — it's an OS designed for human-agent collaboration. safer, more elegant, simpler. go home, hand your machine to it, sleep well. comment rt, giving away 50 invite codes.

macOS is now agent-native

this is crazy. 🤯🤯🤯 in 🍌2 you can generate images with aspect ratios 8:1 and 1:8, and combined with flowith's infinite canvas batch gen, you get this... comment rt, giving away 30 monthly memberships! free to generate with all the other models too

it's 2026. if you're still typing into a chatbox like it's 2024, i don't know what to tell you. the canvas is the interface now. generate anything, anywhere, at insane quality. just play.