Instructions to use AtomGradient/Qwen3.5-9B-Chronos-4bit with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use AtomGradient/Qwen3.5-9B-Chronos-4bit with MLX:

# Make sure mlx-lm is installed
# pip install --upgrade mlx-lm

# Generate text with mlx-lm
from mlx_lm import load, generate

model, tokenizer = load("AtomGradient/Qwen3.5-9B-Chronos-4bit")

prompt = "Write a story about Einstein"
messages = [{"role": "user", "content": prompt}]
prompt = tokenizer.apply_chat_template(
    messages, add_generation_prompt=True
)

text = generate(model, tokenizer, prompt=prompt, verbose=True)

Notebooks
Google Colab
Kaggle
Local Apps Settings
LM Studio

How to use AtomGradient/Qwen3.5-9B-Chronos-4bit with Pi:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "AtomGradient/Qwen3.5-9B-Chronos-4bit"

Configure the model in Pi

# Install Pi:
npm install -g @earendil-works/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "mlx-lm": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "AtomGradient/Qwen3.5-9B-Chronos-4bit"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

MLX LM

How to use AtomGradient/Qwen3.5-9B-Chronos-4bit with MLX LM:

Generate or start a chat session

# Install MLX LM
uv tool install mlx-lm
# Interactive chat REPL
mlx_lm.chat --model "AtomGradient/Qwen3.5-9B-Chronos-4bit"

Run an OpenAI-compatible server

# Install MLX LM
uv tool install mlx-lm
# Start the server
mlx_lm.server --model "AtomGradient/Qwen3.5-9B-Chronos-4bit"
# Calling the OpenAI-compatible server with curl
curl -X POST "http://localhost:8000/v1/chat/completions" \
   -H "Content-Type: application/json" \
   --data '{
     "model": "AtomGradient/Qwen3.5-9B-Chronos-4bit",
     "messages": [
       {"role": "user", "content": "Hello"}
     ]
   }'

Hermes Agent

How to use AtomGradient/Qwen3.5-9B-Chronos-4bit with Hermes Agent:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "AtomGradient/Qwen3.5-9B-Chronos-4bit"

Configure Hermes

# Install Hermes:
curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash
hermes setup
# Point Hermes at the local server:
hermes config set model.provider custom
hermes config set model.base_url http://127.0.0.1:8080/v1
hermes config set model.default AtomGradient/Qwen3.5-9B-Chronos-4bit

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use AtomGradient/Qwen3.5-9B-Chronos-4bit with OpenClaw:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "AtomGradient/Qwen3.5-9B-Chronos-4bit"

Configure OpenClaw

# Install OpenClaw:
npm install -g openclaw@latest
# Register the local server and set it as the default model:
openclaw onboard --non-interactive --mode local \
  --auth-choice custom-api-key \
  --custom-base-url http://127.0.0.1:8080/v1 \
  --custom-model-id "AtomGradient/Qwen3.5-9B-Chronos-4bit" \
  --custom-provider-id mlx-lm \
  --custom-compatibility openai \
  --custom-text-input \
  --accept-risk \
  --skip-health

Run OpenClaw

openclaw agent --local --agent main --message "Hello from Hugging Face"

Qwen3.5-9B-Chronos-4bit / README.md

ShuhongWu

Initial model release

83c7e3f 8 days ago

preview code

Raw History Blame Contribute Delete

2.52 kB

metadata

license: apache-2.0
library_name: mlx
tags:
  - mlx
  - continual-learning
  - on-device-ai
  - neural-imprint
  - chronos
pipeline_tag: text-generation

AtomGradient/Qwen3.5-9B-Chronos-4bit

中文简介

Qwen3.5-9B-Chronos-4bit 是专为 Neural Imprint Chronos 打造的端侧原生持续学习（Native Continual Learning）专用基座模型，需搭载 Chronos 专用内核（Chronos Kernel） 运行。

核心特性

端侧原生持续学习：无需云端训练或反向传播微调，在本地设备运行过程中即可实时吸收、更新与巩固用户的长期偏好、生活习惯、跨事件因果关联及专属工具调用协议。
恒定体积记忆与跨端流转：持续学习产生的用户记忆以恒定超紧凑状态包形式持久化，不随交互轮次线性膨胀，支持在手机、桌面工作站、智能座舱与具身终端之间秒级迁移与恢复。
数据不出端与用户资产主权：所有学习、存储与推理全程在用户本地设备完成；状态资产受机主与内核双重主权保护，确保个人记忆数据的绝对私密性。

使用说明：本模型与 Neural Imprint Chronos 专用内核 深度协同耦合，不支持脱离 Chronos 内核直接使用通用推理框架运行。

English Overview

Qwen3.5-9B-Chronos-4bit is a dedicated on-device backbone model purpose-built for Neural Imprint Chronos native continual learning. It requires the Chronos Kernel runtime to operate.

Key Capabilities

On-Device Native Continual Learning: Continuously learns, updates, and consolidates user preferences, temporal habits, multi-event causal context, and custom tool-calling protocols in real time on edge devices—without cloud retraining or gradient backpropagation.
Constant-Envelope Memory & Cross-Device Mobility: Continual learning states are persisted in an ultra-compact, constant-size state bundle with zero byte growth over time, enabling instant cold-start restoration and seamless migration across smartphones, desktops, vehicle cockpits, and embodied robotics.
Strict On-Device Privacy & User Sovereignty: User data physically never leaves the device. Learned memory states are cryptographically bound to the device owner and the Chronos runtime.

Runtime Requirement: This model is tightly coupled with the Neural Imprint Chronos Kernel and is not intended for standalone execution on generic inference runtimes without the Chronos kernel.