Instructions to use saviochow/GLM-4.6-REAP-268B-A32B-mlx-2Bit with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use saviochow/GLM-4.6-REAP-268B-A32B-mlx-2Bit with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="saviochow/GLM-4.6-REAP-268B-A32B-mlx-2Bit")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("saviochow/GLM-4.6-REAP-268B-A32B-mlx-2Bit")
model = AutoModelForCausalLM.from_pretrained("saviochow/GLM-4.6-REAP-268B-A32B-mlx-2Bit", device_map="auto")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

MLX

How to use saviochow/GLM-4.6-REAP-268B-A32B-mlx-2Bit 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("saviochow/GLM-4.6-REAP-268B-A32B-mlx-2Bit")

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

vLLM

How to use saviochow/GLM-4.6-REAP-268B-A32B-mlx-2Bit with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "saviochow/GLM-4.6-REAP-268B-A32B-mlx-2Bit"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "saviochow/GLM-4.6-REAP-268B-A32B-mlx-2Bit",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/saviochow/GLM-4.6-REAP-268B-A32B-mlx-2Bit

SGLang

How to use saviochow/GLM-4.6-REAP-268B-A32B-mlx-2Bit with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "saviochow/GLM-4.6-REAP-268B-A32B-mlx-2Bit" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "saviochow/GLM-4.6-REAP-268B-A32B-mlx-2Bit",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "saviochow/GLM-4.6-REAP-268B-A32B-mlx-2Bit" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "saviochow/GLM-4.6-REAP-268B-A32B-mlx-2Bit",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

How to use saviochow/GLM-4.6-REAP-268B-A32B-mlx-2Bit with Pi:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "saviochow/GLM-4.6-REAP-268B-A32B-mlx-2Bit"

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": "saviochow/GLM-4.6-REAP-268B-A32B-mlx-2Bit"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

MLX LM

How to use saviochow/GLM-4.6-REAP-268B-A32B-mlx-2Bit with MLX LM:

Generate or start a chat session

# Install MLX LM
uv tool install mlx-lm
# Interactive chat REPL
mlx_lm.chat --model "saviochow/GLM-4.6-REAP-268B-A32B-mlx-2Bit"

Run an OpenAI-compatible server

# Install MLX LM
uv tool install mlx-lm
# Start the server
mlx_lm.server --model "saviochow/GLM-4.6-REAP-268B-A32B-mlx-2Bit"
# Calling the OpenAI-compatible server with curl
curl -X POST "http://localhost:8000/v1/chat/completions" \
   -H "Content-Type: application/json" \
   --data '{
     "model": "saviochow/GLM-4.6-REAP-268B-A32B-mlx-2Bit",
     "messages": [
       {"role": "user", "content": "Hello"}
     ]
   }'

Docker Model Runner
How to use saviochow/GLM-4.6-REAP-268B-A32B-mlx-2Bit with Docker Model Runner:
```
docker model run hf.co/saviochow/GLM-4.6-REAP-268B-A32B-mlx-2Bit
```

Hermes Agent

How to use saviochow/GLM-4.6-REAP-268B-A32B-mlx-2Bit 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 "saviochow/GLM-4.6-REAP-268B-A32B-mlx-2Bit"

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 saviochow/GLM-4.6-REAP-268B-A32B-mlx-2Bit

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use saviochow/GLM-4.6-REAP-268B-A32B-mlx-2Bit with OpenClaw:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "saviochow/GLM-4.6-REAP-268B-A32B-mlx-2Bit"

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 "saviochow/GLM-4.6-REAP-268B-A32B-mlx-2Bit" \
  --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"

How good is it?

by thejson - opened Nov 11, 2025

Discussion

thejson

Nov 11, 2025

Since this is 2bit, was wondering if you tested and if it's stable for long context (eg. 64K or more). Trying to fit in 96GB Ultra and there don't seem to be good models in mlx.

saviochow

Owner Nov 16, 2025

For long context, no. And honestly would not recommend this as it thinks for a long time, outputs something, and seemingly starts thinking again and goes into loops.

saviochow

Owner Nov 16, 2025

GLM 4.6 gguf from unsloth actually worked better and fits into 96GB but at a smaller context, still waiting for GLM 4.6 Air.

saviochow

Owner Nov 16, 2025

Also check out mradermacher/MiniMax-M2-THRIFT-i1-GGUF i1-IQ3_M, hope it helps.

thejson

Nov 16, 2025

•

edited Nov 16, 2025

GLM 4.6 gguf from unsloth actually worked better and fits into 96GB but at a smaller context, still waiting for GLM 4.6 Air.

GGUF seems slower than MLX and trying to stick to MLX if possible. But I'll check out MiniMax-M2 from mradermacher. Agree GLM 4.6 Air would be good middle ground for quality and speed once it's released.

Have you tried any other REAP models?

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment