Instructions to use apus-ailab/APUS-OpenJev-v1-9B-MLX-8bit with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use apus-ailab/APUS-OpenJev-v1-9B-MLX-8bit 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("apus-ailab/APUS-OpenJev-v1-9B-MLX-8bit")

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 apus-ailab/APUS-OpenJev-v1-9B-MLX-8bit with Pi:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "apus-ailab/APUS-OpenJev-v1-9B-MLX-8bit"

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": "apus-ailab/APUS-OpenJev-v1-9B-MLX-8bit"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

MLX LM

How to use apus-ailab/APUS-OpenJev-v1-9B-MLX-8bit with MLX LM:

Generate or start a chat session

# Install MLX LM
uv tool install mlx-lm
# Interactive chat REPL
mlx_lm.chat --model "apus-ailab/APUS-OpenJev-v1-9B-MLX-8bit"

Run an OpenAI-compatible server

# Install MLX LM
uv tool install mlx-lm
# Start the server
mlx_lm.server --model "apus-ailab/APUS-OpenJev-v1-9B-MLX-8bit"
# Calling the OpenAI-compatible server with curl
curl -X POST "http://localhost:8000/v1/chat/completions" \
   -H "Content-Type: application/json" \
   --data '{
     "model": "apus-ailab/APUS-OpenJev-v1-9B-MLX-8bit",
     "messages": [
       {"role": "user", "content": "Hello"}
     ]
   }'

Hermes Agent

How to use apus-ailab/APUS-OpenJev-v1-9B-MLX-8bit 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 "apus-ailab/APUS-OpenJev-v1-9B-MLX-8bit"

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 apus-ailab/APUS-OpenJev-v1-9B-MLX-8bit

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use apus-ailab/APUS-OpenJev-v1-9B-MLX-8bit with OpenClaw:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "apus-ailab/APUS-OpenJev-v1-9B-MLX-8bit"

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 "apus-ailab/APUS-OpenJev-v1-9B-MLX-8bit" \
  --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"

APUS-OpenJev-v1-9B-MLX-8bit / README.zh-CN.md

gump2049

Add Frozen80 per-question results

4a111c9 verified 9 days ago

preview code

Raw History Blame Contribute Delete

2.86 kB

metadata

library_name: mlx
license: apache-2.0
base_model: apus-ailab/APUS-OpenJev-v1-9B
base_model_relation: quantized
pipeline_tag: text-generation
language:
  - en
  - zh
tags:
  - apus-openjev
  - decision-model
  - mlx
  - apple-silicon

APUS-OpenJev-v1-9B-MLX-8bit

English | 中文 · 源模型 · Collection · GGUF collection · MLX collection · MLX-4bit · GGUF / Ollama

APUS-OpenJev-v1-9B 的 MLX 权重（8bit affine，group size 64），适用于 Apple Silicon Mac（mlx-lm、LM Studio）。

OpenJev 是决策模型：每个请求给出状态、指令和 2–16 个候选，模型对候选标签 A–P 打分。它不是聊天模型。

快速开始

pip install mlx-lm
hf download apus-ailab/APUS-OpenJev-v1-9B-MLX-8bit --local-dir ./openjev
python ./openjev/examples/openjev_mlx.py --model ./openjev

examples/openjev_mlx.py 使用 openjev_contracts.py（训练时的格式）渲染 prompt，返回精确的候选分布。

一致性

Frozen80 使用完全相同的 prompt token，与 HF BF16 发布版（完整深度，**68/80 · 85.00%**）对比：

Run / 运行	Backend / 后端	Frozen80	= HF BF16	Max Δp
NVIDIA RTX PRO 6000 (CUDA)	mlx 0.32.2 on Linux x86_64 (Device(gpu, 0))	69/80 · 86.25%	79/80	0.0622

在 Linux（CUDA）上用 MLX 转换并评测；MLX 文件与平台无关，可直接在 Apple Silicon 上加载（4B 的同类文件在 CUDA 与 Metal 上决策完全一致）。Frozen80 评测峰值内存 11.28 GB，建议使用统一内存不少于 16 GB 的 Mac。Frozen80 是复用的开发面板，不是盲测。逐题明细（候选概率、选择、是否正确，可按 panel_index 与 Frozen80 关联）：evaluation/per-question/。

转换说明

mlx-lm 0.31.3 / mlx 0.32.2；8bit affine，group size 64。
GDN 的 A_log 和 linear_attn.norm.weight 保持源精度（35B 发布版中为 FP32）。
仅完整深度、仅文本，概率未经校准。

许可

Apache-2.0，继承自源模型，见 LICENSE。基座模型：Qwen/Qwen3.5-9B。

作者： gumpcheng（xDAN2099）、zhangxu、APUS AI-LAB。