Instructions to use machinez/Qwen3.6-35B-REAP-Pruned-ratio-0.5 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use machinez/Qwen3.6-35B-REAP-Pruned-ratio-0.5 with Transformers:

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

pipe = pipeline("image-text-to-text", model="machinez/Qwen3.6-35B-REAP-Pruned-ratio-0.5")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
pipe(text=messages)

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("machinez/Qwen3.6-35B-REAP-Pruned-ratio-0.5", device_map="auto")

llama-cpp-python

How to use machinez/Qwen3.6-35B-REAP-Pruned-ratio-0.5 with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="machinez/Qwen3.6-35B-REAP-Pruned-ratio-0.5",
	filename="qwen3.6-35b-reap-Q3_K_M.gguf",
)

llm.create_chat_completion(
	messages = [
		{
			"role": "user",
			"content": [
				{
					"type": "text",
					"text": "Describe this image in one sentence."
				},
				{
					"type": "image_url",
					"image_url": {
						"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
					}
				}
			]
		}
	]
)

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use machinez/Qwen3.6-35B-REAP-Pruned-ratio-0.5 with llama.cpp:

Install (macOS, Linux)

curl -LsSf https://llama.app/install.sh | sh
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf machinez/Qwen3.6-35B-REAP-Pruned-ratio-0.5:Q3_K_M
# Run inference directly in the terminal:
llama cli -hf machinez/Qwen3.6-35B-REAP-Pruned-ratio-0.5:Q3_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf machinez/Qwen3.6-35B-REAP-Pruned-ratio-0.5:Q3_K_M
# Run inference directly in the terminal:
llama cli -hf machinez/Qwen3.6-35B-REAP-Pruned-ratio-0.5:Q3_K_M

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf machinez/Qwen3.6-35B-REAP-Pruned-ratio-0.5:Q3_K_M
# Run inference directly in the terminal:
./llama-cli -hf machinez/Qwen3.6-35B-REAP-Pruned-ratio-0.5:Q3_K_M

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf machinez/Qwen3.6-35B-REAP-Pruned-ratio-0.5:Q3_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf machinez/Qwen3.6-35B-REAP-Pruned-ratio-0.5:Q3_K_M

Use Docker

docker model run hf.co/machinez/Qwen3.6-35B-REAP-Pruned-ratio-0.5:Q3_K_M

LM Studio
Jan

vLLM

How to use machinez/Qwen3.6-35B-REAP-Pruned-ratio-0.5 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "machinez/Qwen3.6-35B-REAP-Pruned-ratio-0.5"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "machinez/Qwen3.6-35B-REAP-Pruned-ratio-0.5",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Use Docker

docker model run hf.co/machinez/Qwen3.6-35B-REAP-Pruned-ratio-0.5:Q3_K_M

SGLang

How to use machinez/Qwen3.6-35B-REAP-Pruned-ratio-0.5 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 "machinez/Qwen3.6-35B-REAP-Pruned-ratio-0.5" \
    --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": "machinez/Qwen3.6-35B-REAP-Pruned-ratio-0.5",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

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 "machinez/Qwen3.6-35B-REAP-Pruned-ratio-0.5" \
        --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": "machinez/Qwen3.6-35B-REAP-Pruned-ratio-0.5",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Ollama
How to use machinez/Qwen3.6-35B-REAP-Pruned-ratio-0.5 with Ollama:
```
ollama run hf.co/machinez/Qwen3.6-35B-REAP-Pruned-ratio-0.5:Q3_K_M
```

Unsloth Studio

How to use machinez/Qwen3.6-35B-REAP-Pruned-ratio-0.5 with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for machinez/Qwen3.6-35B-REAP-Pruned-ratio-0.5 to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for machinez/Qwen3.6-35B-REAP-Pruned-ratio-0.5 to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for machinez/Qwen3.6-35B-REAP-Pruned-ratio-0.5 to start chatting

How to use machinez/Qwen3.6-35B-REAP-Pruned-ratio-0.5 with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf machinez/Qwen3.6-35B-REAP-Pruned-ratio-0.5:Q3_K_M

Configure the model in Pi

# Install Pi:
npm install -g @mariozechner/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "machinez/Qwen3.6-35B-REAP-Pruned-ratio-0.5:Q3_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use machinez/Qwen3.6-35B-REAP-Pruned-ratio-0.5 with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf machinez/Qwen3.6-35B-REAP-Pruned-ratio-0.5:Q3_K_M

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 machinez/Qwen3.6-35B-REAP-Pruned-ratio-0.5:Q3_K_M

Run Hermes

hermes

Atomic Chat new

OpenClaw new

How to use machinez/Qwen3.6-35B-REAP-Pruned-ratio-0.5 with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf machinez/Qwen3.6-35B-REAP-Pruned-ratio-0.5:Q3_K_M

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 "machinez/Qwen3.6-35B-REAP-Pruned-ratio-0.5:Q3_K_M" \
  --custom-provider-id llama-cpp \
  --custom-compatibility openai \
  --custom-text-input \
  --accept-risk \
  --skip-health

Run OpenClaw

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

Docker Model Runner
How to use machinez/Qwen3.6-35B-REAP-Pruned-ratio-0.5 with Docker Model Runner:
```
docker model run hf.co/machinez/Qwen3.6-35B-REAP-Pruned-ratio-0.5:Q3_K_M
```

Lemonade

How to use machinez/Qwen3.6-35B-REAP-Pruned-ratio-0.5 with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull machinez/Qwen3.6-35B-REAP-Pruned-ratio-0.5:Q3_K_M

Run and chat with the model

lemonade run user.Qwen3.6-35B-REAP-Pruned-ratio-0.5-Q3_K_M

List all available models

lemonade list

Qwen3.6-35B-A3B REAP Pruned Ratio 0.5

This model was converted to GGUF format from RangerX/Qwen3.6-35B-REAP-Pruned-ratio-0.5 using llama.cpp's llama-imatrix & llama-quantize. It only has Q3_K_M. For Q4_K_M see https://huggingface.co/lennyhans/Qwen3.6-35B-REAP-Pruned-ratio-0.5-Q4_K_M-GGUF.

This repository contains a REAP-pruned version of Qwen/Qwen3.6-35B-A3B. The checkpoint was produced with routed-expert pruning using REAP (Router-weighted Expert Activation Pruning), which scores routed experts with router weights and expert activation norms.

Pruning Settings

Setting	Value
Base model	`Qwen/Qwen3.6-35B-A3B`
Compression / pruning ratio	`0.50`
Pruning method	`reap`
Calibration samples	`1024`
Calibration sequence length	`2048`
Seed	`42`
Router weight renormalization	`true`
Routed experts per MoE layer	`256 -> 128`
Routed experts selected per token	`8`
Shared experts	Preserved
Precision	BF16
Quantization	None

Calibration Data

The calibration set used the REAP paper/code mixture with 1024 total samples:

theblackcat102/evol-codealpaca-v1: 171 samples
Salesforce/xlam-function-calling-60k: 171 samples
open-r1/Mixture-of-Thoughts[code]: 171 samples
open-r1/Mixture-of-Thoughts[math]: 171 samples
open-r1/Mixture-of-Thoughts[science]: 170 samples
SWE-bench/SWE-smith-trajectories(tool): 170 samples

Integration Notes

This checkpoint was generated with packed Qwen3.5/Qwen3.6 REAP support. The packed routed expert tensors and router rows were sliced while preserving the shared expert and the vision-language configuration. The saved model uses the Transformers qwen3_5_moe architecture and includes tokenizer and processor files.

Citation

@inproceedings{
    lasby2026reap,
    title={{REAP} the Experts: Why Pruning Prevails for One-Shot MoE compression},
    author={Mike Lasby and Ivan Lazarevich and Nish Sinnadurai and Sean Lie and Yani Ioannou and Vithursan Thangarasa},
    booktitle={The Fourteenth International Conference on Learning Representations},
    year={2026},
    url={https://openreview.net/forum?id=ukGxWd2aDG}
}

Downloads last month: 127

GGUF

Model size

19B params

Architecture

qwen35moe

Hardware compatibility

3-bit

Model tree for machinez/Qwen3.6-35B-REAP-Pruned-ratio-0.5

Base model

Qwen/Qwen3.6-35B-A3B

Finetuned

RangerX/Qwen3.6-35B-REAP-Pruned-ratio-0.5

Quantized

(4)

this model