Run 35B MoE models in ~2.5 GB of RAM. Weights stream from SSD; a trained router predicts loads a step ahead. Mac, iPhone, Android, Windows supported
Python
README: banner image, Hugging Face badges and checkpoint links
linyubupadd2cec85f078
diff --git a/README.md b/README.mdindex 385de7a..44150b7 100644--- a/README.md+++ b/README.md@@ -1,6 +1,19 @@+<div align="center">++<img src="assets/20260908-223115.jpg" alt="edge0" width="100%">+# edge0-[English](README.md) | [中文](README_zh.md)+**An open-source streaming MoE inference framework — SSD expert offload + parallel LoRA + prerouter routing prediction.**++[](https://huggingface.co/Edge0/Edge0-35b-a3b-preview)+[](https://huggingface.co/Edge0/Edge0-10b-a1b-preview)+[](https://github.com/Edge0-AI/edge0)+[](LICENSE)++English | [中文](README_zh.md)++</div>**edge0** is an open-source streaming MoE inference framework. Itgeneralizes the production-proven recipe — **SSD expert offload +@@ -15,8 +28,8 @@ trained prerouter heads work together as one unit.| Tier | Released checkpoint | Inference profile ||---|---|---|-| `edge0-35b` | `Edge0/Edge0-35b-a3b-preview` | 4-bit, 40 layers, 256 experts, prerouter K=4 |-| `edge0-10b` | `Edge0/Edge0-10b-a1b-preview` | 4-bit, 24 layers, 128 experts, prerouter K=8 |+| `edge0-35b` | [`Edge0/Edge0-35b-a3b-preview`](https://huggingface.co/Edge0/Edge0-35b-a3b-preview) | 4-bit, 40 layers, 256 experts, prerouter K=4 |+| `edge0-10b` | [`Edge0/Edge0-10b-a1b-preview`](https://huggingface.co/Edge0/Edge0-10b-a1b-preview) | 4-bit, 24 layers, 128 experts, prerouter K=8 |Both checkpoints are built on open sparse-MoE base models (Qwen3.5-MoE35B-A3B and the Ling 3.0 bailing hybrid respectively) and ship with thediff --git a/README_zh.md b/README_zh.mdindex 06ce4e1..ea33a25 100644--- a/README_zh.md+++ b/README_zh.md@@ -1,7 +1,20 @@+<div align="center">++<img src="assets/20260908-223115.jpg" alt="edge0" width="100%">+# edge0+**开源流式 MoE 推理框架 —— SSD 专家 offload + 并行 LoRA + prerouter 路由预判**++[](https://huggingface.co/Edge0/Edge0-35b-a3b-preview)+[](https://huggingface.co/Edge0/Edge0-10b-a1b-preview)+[](https://github.com/Edge0-AI/edge0)+[](LICENSE)+[English](README.md) | 中文+</div>+**edge0** 是一个开源的流式 MoE 推理框架:把「SSD 专家 offload +并行 LoRA + prerouter 路由预判」抽象成可扩展的通用框架。后端隔离设计,当前实现 MLX 后端(Apple Silicon),更多平台(CUDA 等)即将接入。@@ -11,8 +24,8 @@| 档位 | 发布 checkpoint | 推理档 ||---|---|---|-| `edge0-35b` | `Edge0/Edge0-35b-a3b-preview` | 4bit,40 层,256 专家,prerouter K=4 |-| `edge0-10b` | `Edge0/Edge0-10b-a1b-preview` | 4bit,24 层,128 专家,prerouter K=8 |+| `edge0-35b` | [`Edge0/Edge0-35b-a3b-preview`](https://huggingface.co/Edge0/Edge0-35b-a3b-preview) | 4bit,40 层,256 专家,prerouter K=4 |+| `edge0-10b` | [`Edge0/Edge0-10b-a1b-preview`](https://huggingface.co/Edge0/Edge0-10b-a1b-preview) | 4bit,24 层,128 专家,prerouter K=8 |两个 checkpoint 均基于开源稀疏 MoE 基座(分别为 Qwen3.5-MoE 35B-A3B与 Ling 3.0 混合架构),并携带为本框架训练的 LoRA 与 prerouter 权重——diff --git a/assets/20260908-223115.jpg b/assets/20260908-223115.jpgnew file mode 100644index 0000000..351116dBinary files /dev/null and b/assets/20260908-223115.jpg differ