---
license: apache-2.0
library_name: mlx
tags:
- moe
- edge-inference
- prerouter
- lora
- ssd-offload
base_model:
- inclusionAI/Ling-3.0-tiny-base
pipeline_tag: text-generation
---

<div align="center">

<img src="20260908-223115.jpg" alt="edge0" width="100%">

# Edge0-8b-a1b Preview

**An 8B-class sparse MoE that runs on a phone — 1.0 GiB of active memory, experts streamed from SSD.**

[![GitHub](https://img.shields.io/badge/GitHub-Edge0--AI%2Fedge0-black?style=for-the-badge&logo=github)](https://github.com/Edge0-AI/edge0)
[![Hugging Face](https://img.shields.io/badge/%F0%9F%A4%97%20Hugging%20Face-Edge0--35b--a3b--preview-yellow?style=for-the-badge)](https://huggingface.co/Edge0/Edge0-35b-a3b-preview)
[![Hugging Face](https://img.shields.io/badge/%F0%9F%A4%97%20Hugging%20Face-Edge0--8b--a1b--preview-yellow?style=for-the-badge)](https://huggingface.co/Edge0/Edge0-8b-a1b-preview)
[![License](https://img.shields.io/badge/License-Apache%202.0-blue?style=for-the-badge)](https://github.com/Edge0-AI/edge0/blob/main/LICENSE)

</div>

**Edge0-8b-a1b** — an 8B MoE LLM that runs at viable speed on portable devices in under **1.0 GiB of active memory** (1/4 of its 4.2 GB weight footprint), 
via the [edge0](https://github.com/Edge0-AI/edge0) streaming inference framework.

The key is streaming: experts are memory-mapped and fetched from SSD only as routed, 
so RAM holds just the active weights. What makes that viable — instead of stalling like plain parameter offloading — 
is a trained prerouter head that **predicts the next token's expert routing one step ahead**, hiding storage latency behind compute.


> **Preview status:** this is an early preview release of the edge0
> pipeline. The checkpoint ships as int4 quantization plus LoRA and
> prerouter adapters trained for this framework.

## Model summary

| | |
|---|---|
| Base model | inclusionAI Ling 3.0 tiny (bailing hybrid, MLA + MoE, ≈7.9B total / ≈1.2B active) |
| Quantization | 4-bit |
| Layers | 24 |
| Experts / active per token | 128 / 8 (K=8) |
| Framework | [edge0](https://github.com/Edge0-AI/edge0) (MLX backend) |
| Contents | base checkpoint + `lora_edge0_8b.safetensors` + `prerouter_edge0_8b.safetensors` |

The LoRA and prerouter adapters are co-located with the base checkpoint
and load automatically — this repository is a complete, ready-to-run
model directory for `edge0`.

## Quality

Both models were evaluated by us under identical settings and
parameters. The loss of the edge0 pipeline (int4 + adapters) relative to
the fp16 base model is small: **2.8 points on average**, with MMLU-Pro
above the base (max 100):

| Benchmark | edge0-8b (int4) | Ling 3.0 tiny (fp16) |
|---|---:|---:|
| AIME 2026 | 63.3 | 73.3 |
| HumanEval | 91.5 | 92.7 |
| GPQA-Diamond | 70.7 | 71.2 |
| MMLU-Pro | 70.1 | 65.8 |
| IFBench | 53.9 | 60.6 |
| **Average** | **69.9** | **72.7** |

## Performance

Measured with `examples/bench.py` on a Mac mini M4 Pro, 24 GB:

| Decode speed | Prefill throughput (cold / warm) | Peak active memory* |
|---|---|---|
| 23.9–25.3 tok/s | 500 / 1428 tok/s | 1.0 GiB |

*Short contexts; long contexts add KV cache (≈3.3 GiB at 3.3k tokens).
Expert weights stream from SSD via mmap and are not resident.

## Quick start

```bash
pip install -e 'git+https://github.com/Edge0-AI/edge0.git#egg=edge0[fetch]'

# Download this repository into a local directory
huggingface-cli download Edge0/Edge0-8b-a1b-preview --local-dir ./Edge0-8b-a1b-preview

# Run it
export EDGE0_8B_MODEL=$PWD/Edge0-8b-a1b-preview
edge0 chat --name edge0-8b --prompt "Introduce yourself"

# Or serve an OpenAI-compatible HTTP API
edge0 serve --name edge0-8b --port 8083
```

For full usage (Python API, streaming options, prerouter details), see the
[edge0 documentation](https://github.com/Edge0-AI/edge0#documentation).

## License

Apache 2.0. See [LICENSE](https://github.com/Edge0-AI/edge0/blob/main/LICENSE).