---
library_name: aneforge
license: mit
base_model: openai-community/gpt2
pipeline_tag: text-generation
tags:
- aneforge
- apple-neural-engine
---

# GPT-2 (ANEForge)

An unmodified duplicate of [`openai-community/gpt2`](https://huggingface.co/openai-community/gpt2), tagged for use with
[**ANEForge**](https://github.com/sbryngelson/ANEForge) so the weights load and run directly on the
Apple Neural Engine (no CoreML). Weights are byte-identical to the source; see the original repo for
the model details. Licensed `mit` from the source.

## Use with ANEForge

```python
# pip install "aneforge[models]"
import aneforge as af
from transformers import AutoTokenizer

tok = AutoTokenizer.from_pretrained("aneforge/gpt2")
model = af.load_llm("aneforge/gpt2")            # prefill + resident-KV-cache decode on the ANE
ids = tok.encode("The Neural Engine is")
print(tok.decode(model.generate(ids, max_new_tokens=20)))
```

ANEForge compiles the model's graph into a single ANE program and streams the weights from this repo
via `huggingface_hub`. See the [docs](https://aneforge.readthedocs.io) and the
[paper](https://arxiv.org/abs/2606.17090).