Instructions to use zenlm/zen4-mini with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use zenlm/zen4-mini with Transformers:

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

pipe = pipeline("text-generation", model="zenlm/zen4-mini")
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 AutoProcessor, AutoModelForImageTextToText

processor = AutoProcessor.from_pretrained("zenlm/zen4-mini")
model = AutoModelForImageTextToText.from_pretrained("zenlm/zen4-mini")
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?"}
        ]
    },
]
inputs = processor.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(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use zenlm/zen4-mini with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "zenlm/zen4-mini"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "zenlm/zen4-mini",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/zenlm/zen4-mini

SGLang

How to use zenlm/zen4-mini 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 "zenlm/zen4-mini" \
    --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": "zenlm/zen4-mini",
		"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 "zenlm/zen4-mini" \
        --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": "zenlm/zen4-mini",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use zenlm/zen4-mini with Docker Model Runner:
```
docker model run hf.co/zenlm/zen4-mini
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

Zen4 Mini

Zen4 Mini is a 4B parameter language model from the Zen4 family by Zen LM and Hanzo AI.

Built on open-weight weights with Zen4 Frontier architecture for unrestricted, open-ended AI assistance.

Model Details

Property	Value
Parameters	4B total, 4B active
Architecture	Zen4 Frontier
Context	262K tokens
License	APACHE-2.0
Family	Zen4
Tier	Small
Creator	Zen LM / Hanzo AI

Usage

from transformers import AutoModelForCausalLM, AutoTokenizer

model = AutoModelForCausalLM.from_pretrained("zenlm/zen4-mini", torch_dtype="auto")
tokenizer = AutoTokenizer.from_pretrained("zenlm/zen4-mini")

messages = [{"role": "user", "content": "Hello, who are you?"}]
text = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
inputs = tokenizer(text, return_tensors="pt").to(model.device)
outputs = model.generate(**inputs, max_new_tokens=512)
print(tokenizer.decode(outputs[0][inputs.input_ids.shape[-1]:], skip_special_tokens=True))

Zen4 Family

Model	Parameters	Context	HuggingFace
Zen4 Nano	0.8B	262K	zenlm/zen4-nano
Zen4 Micro	2B	262K	zenlm/zen4-micro
Zen4 Mini	4B	262K	zenlm/zen4-mini
Zen4	9B	262K	zenlm/zen4
Zen4 Pro	27B	262K	zenlm/zen4-pro
Zen4 Max	35B MoE (3B active)	262K	zenlm/zen4-max
Zen4 Coder Flash	31B MoE (3B active)	131K	zenlm/zen4-coder-flash
Zen4 Pro Max	80B MoE (3B active)	256K	zenlm/zen4-pro-max
Zen4 Coder	80B MoE (3B active)	256K	zenlm/zen4-coder
Zen4 Mega	122B MoE (10B active)	262K	zenlm/zen4-mega
Zen4 Thunder	230B MoE (10B active)	1M	zenlm/zen4-thunder
Zen4 Storm	456B MoE (45B active)	1M	zenlm/zen4-storm
Zen4 Titan	744B MoE (40B active)	128K	zenlm/zen4-titan
Zen4 Ultra	1.04T MoE (32B active)	256K	zenlm/zen4-ultra
Zen4 Ultra Max	1T MoE (50B active)	128K	zenlm/zen4-ultra-max

Model tree for zenlm/zen4-mini

Base model

Qwen/Qwen3.5-4B-Base

Finetuned

Qwen/Qwen3.5-4B

Finetuned

(215)

this model

Quantizations

2 models

Duplicated from Qwen/Qwen3.5-4B

zenlm
/

zen4-mini

Zen4 Mini

Model Details

Usage

Zen4 Family

Links

Model tree for zenlm/zen4-mini