Instructions to use yasserrmd/AgentUX-4B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use yasserrmd/AgentUX-4B with Transformers:

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

pipe = pipeline("text-generation", model="yasserrmd/AgentUX-4B")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("yasserrmd/AgentUX-4B")
model = AutoModelForCausalLM.from_pretrained("yasserrmd/AgentUX-4B")
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]:]))

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use yasserrmd/AgentUX-4B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "yasserrmd/AgentUX-4B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "yasserrmd/AgentUX-4B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/yasserrmd/AgentUX-4B

SGLang

How to use yasserrmd/AgentUX-4B 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 "yasserrmd/AgentUX-4B" \
    --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": "yasserrmd/AgentUX-4B",
		"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 "yasserrmd/AgentUX-4B" \
        --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": "yasserrmd/AgentUX-4B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use yasserrmd/AgentUX-4B with Docker Model Runner:
```
docker model run hf.co/yasserrmd/AgentUX-4B
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

🧠 AgentUX‑4B

AgentUX‑4B is a compact, agentic reasoning model designed for UI layout generation, component reasoning, and lightweight code structuring tasks. It’s a 4B-parameter model merged using SLERP (Spherical Linear Interpolation) via MergeKit, combining:

🔷 60% Tesslate/UIGEN-X-4B-0729 — excellent at UI understanding and structured generation
🔹 40% Menlo/Jan-nano — strong generalist with compact tool-use and agentic reasoning

✨ Highlights

📐 UI reasoning & layout structure understanding
🧩 Component-to-code generation (HTML, JSX, CSS fragments)
🧠 Compact agentic planning and multi-step reasoning
⚡ Lightweight & merge-optimized for local inference and real-time apps
🧬 Merged using SLERP to preserve semantic smoothness between sources

🧪 Example Use Cases

Prompt	Task
"Generate a signup form layout using HTML and CSS"	Frontend layout generation
"Explain the role of `flex-wrap` in UI design"	UI reasoning
"Plan 3 steps to build a sidebar menu using React"	Agentic decomposition

🔧 Usage Example

from transformers import AutoTokenizer, AutoModelForCausalLM, pipeline

model_id = "yasserrmd/AgentUX-4B"

tokenizer = AutoTokenizer.from_pretrained(model_id, trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained(model_id, trust_remote_code=True)

pipe = pipeline("text-generation", model=model, tokenizer=tokenizer)

prompt = "Create a responsive layout with sidebar and header using Flexbox."
response = pipe(prompt, max_new_tokens=512)[0]["generated_text"]
print(response)

🛠 Merge Details

🔗 MergeKit method: slerp
🔍 Focused on reasoning alignment between structured generation (UIGEN) and agent-style planning (Jan-nano)
🤖 No additional fine-tuning post-merge

📘 License & Credit

Model licensed under Apache 2.0
All credit to the original base models:
- Tesslate/UIGEN-X-4B-0729
- Menlo/Jan-nano

Downloads last month: 1

Safetensors

Model size

4B params

Tensor type

F16

Model tree for yasserrmd/AgentUX-4B

Menlo/Jan-nano

Tesslate/UIGEN-X-4B-0729

Merge model

this model

Quantizations

3 models