Metadata-Version: 2.1
Name: neuratensor
Version: 1.0.0
Summary: Production neuromorphic inference SDK with CUDA acceleration
Home-page: https://neuramorphic.ai
Author: Neuramorphic, Inc.
Author-email: info@neuramorphic.ai
License: Proprietary
Project-URL: Documentation, https://docs.neuramorphic.ai
Project-URL: Source, https://github.com/neuramorphic/neuratensor-sdk
Project-URL: Bug Reports, https://github.com/neuramorphic/neuratensor-sdk/issues
Keywords: neuromorphic,snn,ssm,cuda,inference,deep-learning,ai
Platform: Linux
Classifier: Development Status :: 4 - Beta
Classifier: Intended Audience :: Developers
Classifier: Intended Audience :: Science/Research
Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
Classifier: License :: Other/Proprietary License
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.8
Classifier: Programming Language :: Python :: 3.9
Classifier: Programming Language :: Python :: 3.10
Classifier: Operating System :: POSIX :: Linux
Classifier: Environment :: GPU :: NVIDIA CUDA
Requires-Python: >=3.8
Description-Content-Type: text/markdown
License-File: LICENSE
Requires-Dist: torch >=2.0.0
Requires-Dist: numpy >=1.19.0
Provides-Extra: cli
Requires-Dist: click >=8.0.0 ; extra == 'cli'
Provides-Extra: dev
Requires-Dist: pytest >=7.0.0 ; extra == 'dev'
Requires-Dist: black >=22.0.0 ; extra == 'dev'
Requires-Dist: flake8 >=4.0.0 ; extra == 'dev'

# NeuraTensor SDK

**Production neuromorphic inference at 8ms latency**

NeuraTensor is a high-performance neuromorphic inference SDK featuring a proprietary fused SNN-SSM architecture with CUDA acceleration.

## ⚡ Performance

- **8.13ms latency** (64M model, Jetson AGX Orin)
- **123 seq/s throughput**
- **120x faster** than PyTorch baseline
- Sub-10ms real-time inference

## �� Quick Start

```bash
pip install neuratensor
```

```python
from neuratensor import NeuraTensor, NeuraTensorConfig

# Load model
config = NeuraTensorConfig.preset("64m")
model = NeuraTensor(config).cuda().half()

# Inference
output = model(input_ids)
# ✅ 8ms latency, 64M params, 123 seq/s
```

## 📊 Models

| Model | Parameters | Latency | Use Case |
|-------|-----------|---------|----------|
| 64M   | 64M | 8ms | Edge, real-time |
| 256M  | 256M | ~15ms | Balanced |
| 1B    | 1B | ~35ms | Quality |

## 🔧 Requirements

- NVIDIA GPU (CUDA 11.4+, SM 7.0+)
- Python 3.8+
- PyTorch 2.0+

## 📚 Documentation

- [Distribution Guide](DISTRIBUTION_GUIDE.md)
- [Patent Notice](PATENT_NOTICE.txt)

## 📄 License

Proprietary. Binary distribution only.

**© 2024-2025 Neuramorphic, Inc.**
