---
license: apache-2.0
library_name: pytorch
pipeline_tag: time-series-forecasting
tags:
- time-series
- forecasting
- pytorch
- safetensors
- deployable
- edge-ai
- onnx
- raspberry-pi
- tiny-ml
- iot
- transformer
- zero-shot
- huggingface
- timesfm-alternative
metrics:
- mase
- smape
---

<h1 align="center">
  🔮 NanoForecast 200k (v0.1)
  <br>
  <sub>World's smallest deployable time series transformer</sub>
</h1>

<p align="center">
  <b>676K params · 256 context · Just 2.7 MB</b>
  <br>
  <i>Fits on a Raspberry Pi Zero. Ships as a dependency, not an API call.</i>
</p>

<p align="center">
  <a href="https://github.com/eulogik/NanoForecast"><img src="https://img.shields.io/badge/GitHub-eulogik%2FNanoForecast-181717?logo=github"></a>
  <a href="https://pypi.org/project/nanoforecast/"><img src="https://img.shields.io/pypi/v/nanoforecast"></a>
  <a href="https://eulogik.com"><img src="https://img.shields.io/badge/by-Eulogik-purple"></a>
</p>

<p align="center">
  💡 Need better accuracy? Check <a href="https://huggingface.co/eulogik/nanoforecast-500k"><b>nanoforecast-500k</b></a> (1.6M params, MASE 3.45) or the latest <a href="https://huggingface.co/eulogik/nanoforecast-v03"><b>nanoforecast-v03</b></a> (6.5M params, MASE 2.73).
</p>

---

## 📦 Quick Start

```bash
pip install nanoforecast
python3 train_from_csv.py --csv my_data.csv --target sales --horizon 48
```

Or use a pretrained model:

```python
from nanoforecast import NanoForecast
model = NanoForecast.from_pretrained("eulogik/nanoforecast-200k")
```

## 📋 When to Use This Model

| Scenario | Pick This Model |
|---|---|
| Raspberry Pi Zero (512 MB RAM) | ✅ **nanoforecast-200k** |
| Browser via ONNX.js (< 1 MB) | ✅ **nanoforecast-200k** |
| AWS Lambda cold starts | ✅ **nanoforecast-200k** |
| Battery-powered IoT sensor | ✅ **nanoforecast-200k** |
| Any accuracy-critical app | ❌ Use [v0.3](https://huggingface.co/eulogik/nanoforecast-v03) instead |

## 🔄 Streaming Inference

Feed one value at a time with DeltaNet stateful RNN:

```python
result = model.predict(context, horizon=48, return_state=True)
state = result.pop("state")
for new_val in incoming_data_stream:
    result = model.predict_step(new_val, state, horizon=48)
```

## 📚 All Variants

| Model | Params | Context | Size | Best For |
|---|---|---|---|---|
| **nanoforecast-200k** (you are here) | **676K** | 256 | 2.7 MB | Extreme edge |
| [nanoforecast-500k](https://huggingface.co/eulogik/nanoforecast-500k) | 1.6M | 256 | 6.4 MB | General purpose |
| [nanoforecast-v03](https://huggingface.co/eulogik/nanoforecast-v03) | 6.5M | 512 | 26 MB | Max accuracy |

## ❤️ Built by Eulogik

[![Eulogik](https://img.shields.io/badge/by-Eulogik-purple)](https://eulogik.com)

**Eulogik** — deployable AI for the real world.

Star the repo ⭐ on [GitHub](https://github.com/eulogik/NanoForecast)!