E-1
Developed by: E Inference Access: Gated — request access required Version: v1.0 (July 2026)
Model Description
E-1 is an 8B parameter frontier model developed by E Inference, designed to run fully on-device and offline on mobile hardware. This is an early, request-access release.
- Developer: E Inference
- Model type: [e.g., decoder-only transformer]
- Parameters: 8B
- Context length: [TBD]
- Quantization / format: [TBD]
Intended Use
This model is intended for local, on-device inference on mobile hardware. It is released on a request-access basis for [research / early partner testing / evaluation purposes — adjust as applicable].
Out of scope: Any use not covered by your approved access request, including redistribution, commercial deployment, or use in training competing models.
How to Get Access
Access to this repository is gated. Click "Request access" above and provide the requested information. Approval is at E Inference's discretion.
How to Use
[Add loading/inference instructions once finalized —
e.g. transformers, llama.cpp, GGUF, MLX, ONNX, etc.]
Limitations
This model may produce inaccurate, incomplete, or biased outputs. It has not been evaluated for safety-critical, medical, legal, or financial use cases. Users are responsible for evaluating outputs before relying on them.
Contact
For licensing, partnership, or access inquiries: [founders@e-inference.com]