README.md

metadata

title: ASI V2.5 ZeroGPU Demo
emoji: 🚀
colorFrom: blue
colorTo: red
sdk: gradio
sdk_version: 4.44.0
app_file: app.py
pinned: false
license: mit

🚀 ASI V2.5: ZeroGPU H200 Performance Demo

REAL GPU ASI V2.5 Testing on NVIDIA H200!

This Space demonstrates the Adaptive Structured Intelligence V2.5 attention mechanism running on ZeroGPU H200 with 70GB VRAM - finally showing the real speedups ASI was designed for!

✨ Features

• 🚀 ZeroGPU H200: NVIDIA H200 with 70GB VRAM (FREE with Pro!)
• 🔥 Real GPU Performance: Actual CUDA optimizations and tensor cores
• 📈 Long Sequences: Test up to 8192 tokens (impossible on CPU)
• ⚡ Mixed Precision: FP16 optimization for H200
• 🎯 Adaptive Attention: Linear O(L) for long sequences, Exact O(L²) for short

🎯 Performance Expectations

Sequence Length	Expected Speedup	Attention Type
1024 tokens	1.2-1.5x	Linear
2048 tokens	1.5-2.0x	Linear
4096 tokens	2.0-2.5x	Linear
8192 tokens	2.5x+	Linear

🔧 How to Use

1. Configure ASI parameters (threshold, feature_dim, etc.)
2. Set sequence lengths to test (try long ones!)
3. Click "Run ZeroGPU ASI V2.5 Test"
4. See REAL GPU speedups!

💡 Why ZeroGPU?

• ✅ FREE with HuggingFace Pro ($9/mois)
• ✅ 70GB VRAM - handles very long sequences
• ✅ NVIDIA H200 - latest architecture
• ✅ 8x quota with Pro priority

🏆 Technical Details

• ASI V2.5: Adaptive attention switching
• CUDA Optimized: Tensor cores + mixed precision
• Memory Efficient: 70GB VRAM utilization
• Production Ready: Real deployment example

📊 Links

• 🤗 Model: khopilot/asi-v25-longformer-core
• 🐙 GitHub: khopilot/asi-v25-longformer-core
• 📦 PyPI: asi-v25-longformer

---

Finally - ASI V2.5 running on real GPU with the speedups it was designed for! 🚀