🏗️ Building on HF

aifeifei

aifeifei798

188 12 638

https://feimatrix.com/

AI & ML interests

roleplay

Recent Activity

liked a model about 20 hours ago

Viggle/Qwen-Image-2.1-viggle-turbo

reacted to SeaWolf-AI's post with 👍 about 20 hours ago

💻 Data-center AI, now on a laptop: POCKET-Darwin-180B We're releasing a 4-bit GGUF build of Darwin-180B-RSI, #1 on seven official Hugging Face leaderboards (self-reported), that runs without a GPU. 📦 360 GB → 111 GB (4-bit GGUF, 4 files) 🖥️ No GPU: one server CPU (16 threads) at 18.4–21.0 tokens/s 💻 RTX 5060 laptop (8 GB VRAM) + 32 GB RAM: 4.17 tokens/s 🧊 128 GB mini PC: whole model in memory, no GPU needed 🎯 MMLU-Pro, 2,000 questions, paired: original 87.65% = 4-bit 87.65% How? · Only ~3B of 180B parameters are active per token (10 of 512 experts) · llama.cpp streams just the needed experts from SSD, so 32 GB RAM is enough · Graft quantization: we took the proven Unsloth UD-Q4_K_XL base build and swapped in only the 300 tensors our RSI training changed (300/300 verified) Under the hood is Model-level Recursive Self-Improvement. The model solves verifiable problems, keeps only its own solutions that check out as correct, and trains on them. No human-written solutions or reasoning traces. Built for teams that can't send data to an external cloud (defense, finance, public sector) to run a top-tier model fully offline. 📝 Article: https://huggingface.co/blog/FINAL-Bench/data-center-ai-now-on-a-laptop-pocket-darwin-180b 🤗 Model: https://huggingface.co/FINAL-Bench/POCKET-Darwin-180B-GGUF 🧬 Original: https://huggingface.co/FINAL-Bench/Darwin-180B-RSI #Darwin #RSI #GGUF #llamacpp #OnDevice #MoE

updated a model about 23 hours ago

aifeifei798/Myriad-MoE-25K-Micro-Experts

View all activity

Organizations

Posts 4

Post

5824

# 🚀 unsloth-gui: 像点外卖一样，训练你自己的AI模型

厌倦了复杂的代码和看不懂的参数？unsloth-gui 让你用最简单的方式，体验 LoRA 微调的全部乐趣。

**核心理念：三步搞定！**

## ✨ 凭什么说这是最简单的？

因为我们把所有复杂的东西都藏起来了。你只需要：

1. **👉 选模型**: 从列表里选一个你喜欢的基础模型，比如 Llama-3, Mistral...
2. **👉 选数据**: 选一份你想让它学习的“剧本”，比如我们内置的“孙悟空”风格数据。
3. **✅ 点开始训练**: 就这样。真的。

**然后？**

喝杯咖啡，回来后你就能在“**测试**”选项卡里，和你亲手训练好的、独一无二的AI模型聊天了。

---

## 🔧 两分钟上手指南

1. **下载 & 安装**

bash
    # 克隆仓库
    git clone https://github.com/aifeifei798/unsloth-gui.git
    cd unsloth-gui

    # 安装依赖
    pip install -r requirements.txt

2. **配置 (可选，可跳过)**
* 想训练自己的模型？编辑 models.json。
* 想用自己的数据？在 datasets_config/ 目录下加一个 .json 文件。
* *...或者，直接用我们内置的孙悟空示例，跳过这一步！*

3. **启动！**

bash
    python app.py

在浏览器里打开 http://127.0.0.1:7860，开始你的“炼丹”之旅！

---

## 🌟 核心功能一览

* **极简训练界面**: 专为“一键启动”设计。
* **实时训练监控**: 内置 TensorBoard，看着你的模型一点点变聪明。
* **即时推理测试**: 训练完立刻就能聊，所见即所得。
* **断点续训**: 训练中断？不怕，接着来。
* **8GB 显卡优化**: 在你的游戏电脑上就能跑！
* **高度可定制**: JSON配置，想加什么模型、什么数据，你说了算。

---

觉得酷？给个 Star ⭐ 吧！