Stephen Cresswell (@Crezza999) header
Stephen Cresswell (@Crezza999) profile picture

Former HS2 Whistleblower & Project Risk Expert

Joined June 2009
Photo by Stephen Cresswell (@Crezza999)Photo by Stephen Cresswell (@Crezza999)Photo by Stephen Cresswell (@Crezza999)Photo by Stephen Cresswell (@Crezza999)Photo by Stephen Cresswell (@Crezza999)
Stephen Cresswell retweeted
Do you feel it? In late 19th-century Paris, while concert halls demanded massive orchestral sound and technical speed, Erik Satie took a radically different path. They made fun of him. They tried to ignore him. Working as a pianist in the cabarets of Montmartre, he composed Gymnopédie No. 1 in 1888—a work that stripped Western classical music to its barest essentials. The piece operates on a deliberate paradox. Though peaceful on the surface, Satie marked it lento e doloroso (slowly and painfully). That underlying melancholy prevents the music from becoming purely decorative, giving every suspended chord weight and quiet gravity. Watching it played, you see the thought and weight put into each simple note as it pulls on something inside you. To many it is not to make you feel good but to feel. And in this is something shared. Perhaps a rainy-day memory. In this 1966 broadcast, French pianist Jacques Février demonstrates the strict discipline Satie’s music requires. Chosen by Maurice Ravel and Francis Poulenc to premiere their works, he understood that performing Satie demands utter emotional restraint, letting the space between the notes carry the composition’s weight. By removing resolving harmonies and traditional momentum, Satie laid the groundwork for modern minimalism and ambient music nearly a century before either genre had a name.
138
588
47
2,942
137,282
Stephen Cresswell retweeted
NVIDIA just unleashed SANA-WM and it’s an absolute MONSTER for the future of open source AI! A blazing-fast 2.6B-parameter open-source world model that doesn’t just generate video… it creates controllable, physics-rich, high-fidelity worlds on demand. Why this is insanely powerful: • One image + text prompt + 6-DoF camera trajectory → generates 720p videos up to 60 seconds long with buttery-smooth, precisely controlled camera movement. You’re not just watching, you’re piloting the simulation. • Runs locally on a single consumer GPU (RTX 5090 level) thanks to heavy distillation + NVFP4 quantization. Full 60-second clip denoised in ~34 seconds. No massive clusters required. • 36× higher throughput than previous open models while rivaling (or beating) closed industrial giants in visual quality and consistency. • Trained lightning-fast: ~213K public videos in just 15 days on 64 H100s. • Built with next-level tech: Hybrid Linear Attention, dual-branch camera control, two-stage pipeline, and rock-solid metric-scale pose understanding. 
This is a true open world model, the foundation for embodied AI, robotics, autonomous systems, and hyper-realistic simulations that can run anywhere. Project: nvlabs.github.io/Sana/WM/
GitHub: github.com/NVlabs/Sana
Paper: arxiv.org/abs/2605.15178 At our Zero-Human Company, we’re already running SANA-WM live in our core pipelines. It’s supercharging autonomous agent training, generating unlimited synthetic training data, and powering full end-to-end simulation loops, zero humans in the loop. The speed and control let us test thousands of edge-case scenarios overnight, iterate at lightspeed, and push our fully autonomous operations further than ever before. This is the kind of breakthrough that turns science fiction into daily reality. World models just leveled up — hard. The age of personal, local, controllable universes is here.
82
315
53
2,478
620,016