Claude Fable 5 is dead and GPT-5.6 delaying launch…
Microsoft has changed the game 🤯
They've open-sourced bitnet.cpp, a 1-bit LLM inference framework.
It let's you run 100B parameter models on your local CPU without GPUs.
- 6.17x faster inference
- 82.2% less energy on CPUs
100% Open Source.
显示更多
You don't need Claude Fable 5 or GPT-5.6
I built with Claude 👇
Drifting particles + pulsing sound rings, the techniques:
1. Flow field = each particle reads a Simplex noise field for which way to drift, organic not random.
2. GPGPU = thousands of particles updated on the GPU in a texture, 60fps on a phone.
3. Audio-reactive shader = the rings pulse to the track frequency data.
4. Selective bloom = only the bright parts glow. Lenis + GSAP + Three.js.
Save this 🔊
显示更多
You don't need Claude Fable 5 or GPT-5.6
I built with Claude 👇
Drifting particles + pulsing sound rings, the techniques:
1. Flow field = each particle reads a Simplex noise field for which way to drift, organic not random.
2. GPGPU = thousands of particles updated on the GPU in a texture, 60fps on a phone.
3. Audio-reactive shader = the rings pulse to the track frequency data.
4. Selective bloom = only the bright parts glow. Lenis + GSAP + Three.js.
Save this 🔊
显示更多