I built a bot on an ESP32 that can create full projects by itself. Ask it to generate images, videos, 3D, build a website and publish it, it does everything autonomously. Powered by Claude Opus 4.6 for code, visual content generated with fal, bot powered via OpenClaw🦞 The skills... to generate project with fal are here All running on a tiny M5Stack Core2.show more

Very good at advicing model

It can generate and show content

@fal @openclaw Great job @OdinLovis ! How hard would it be to make an AI Pin-esque wearable that doesn't necessarily need a screen? Just an audio assistant with vision

@fal @openclaw W lovis

@fal @openclaw SO COOL CAN I HAVE ONE. need a little box i can shake and make cool images/videos with for my kiddo :)

Sorry if I was not clear, the process of sharing information and system is made on the m5stack and openclaw is on a independant vps. The m5stack get the output, process it and show it to user. Also the system is still slow , I want to get the process better.

@gorkem @fal @openclaw fal toys comming soon!

@fal @openclaw Exactement ce que je cherchais ! Merci

@fal @openclaw @OdinLovis would you like to get funding for this?

@fal @openclaw M5 Stack Core 2 is great

@fal @openclaw nice poc. waiting for local models running on esp32 ;)

@fal @openclaw No shot we code works

@fal @openclaw @OdinLovis, that's a fascinating project; using an ESP32 for autonomous project creation is quite impressive.

@fal @openclaw Thanks for sharing - can you share the details on the hardware?

@fal @openclaw Adorable.

@fal @openclaw Autonomous creative agent on a M5Stack is wild. But genuine question: with API calls for Claude + fal generation, doesn't latency kill the experience? Or is this more for overnight batch jobs than real-time creative sessions?

@fal @openclaw Amazing garbage generator.

@fal @openclaw an ESP32 running autonomous creative projects via OpenClaw is wild. the M5Stack form factor means you could have this sitting on a desk just... making things. curious about the latency — does the ESP32 handle the orchestration locally or is it mostly a thin client to the cloud?

@fal @openclaw sounds amazing. would love to learn more actually :)

@fal @openclaw wait youre running opus on an esp32? thats either genius routing or youre burning through api credits faster than i debug retry loops how are you handling state when the bot crashes mid-project? or does openclaw just save checkpoints automatically