Qwen 4.0 (Preview) - just blew my mind and I am speechless.
Massive respect to @QwenDevs and @Alibaba_Qwen
Is this the moment Qwen overtakes Antropic and OpenAI? May be they will release new model soon but I am genuinely excited what we will receive soon especially 27b and flash next.
The open-model race is getting INSANE.
🚨 Fable 5.5 TESTING Part 14
⚡ I asked Claude to ride a lightning bolt from cloud to ground.
One prompt. I never touched the code. 🤯
⏱️ 182 minutes
💸 $17.57 in API costs
🤖 Claude Fable 5.5 Xhigh in Claude Code
What it did on its own 👇
⛈️ Simulated the storm on a 30.5M-cell grid: rising air, rain, ice and electric charge, with 63 flashes logged
🌩️ Grew the bolt with a dielectric breakdown sim: 16 m grid, 53,505 cells, about -110 MV at the root
👁️ Rode the camera on the tip for all 60 steps of the leader
🔥 Burned the scar with the same kind of physics into a prairie of 2.4M grass blades
🎧 Synthesised every sound: the hum climbs with each step, then 130 ms of dead silence before the crack
🐛 Caught the bolt stopping 752 m above the ground and resumed the sim instead of restarting it
🚨 Scanned every frame, found a full-screen strobe at 4 to 6 Hz and fixed it before I ever saw it
📦 28 s · one unbroken shot · 1080p60 · three.js + Taichi
Fable 5.5 just f*cked the whole mecha anime industry...
this is f*cking crazy
i made a giant robot catch a meteor with Fable 5.5, no image generation, no video model, 100% code
> 6 agent teams working in parallel
> an AI art director on every team
> 36 shots, 1,968 code-rendered frames
> zero image generation, zero video models
> every frame coded, not drawn
full workflow + code in the comments
then, read the complete motion design guide
my vocal synth models the vocal tract as a network of branching acoustic tubes, unlike vocaloid which stitches together and pitch-shifts phoneme recordings. still working on improving naturalness by tuning it against recordings of real singers, but I'm already impressed at how cool it sounds!