Instructions to use ZarfixAI/ZarfixAICerdas1.0 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use ZarfixAI/ZarfixAICerdas1.0 with Transformers:

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

pipe = pipeline("text-generation", model="ZarfixAI/ZarfixAICerdas1.0")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("ZarfixAI/ZarfixAICerdas1.0")
model = AutoModelForCausalLM.from_pretrained("ZarfixAI/ZarfixAICerdas1.0", 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=256)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use ZarfixAI/ZarfixAICerdas1.0 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "ZarfixAI/ZarfixAICerdas1.0"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ZarfixAI/ZarfixAICerdas1.0",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/ZarfixAI/ZarfixAICerdas1.0

SGLang

How to use ZarfixAI/ZarfixAICerdas1.0 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 "ZarfixAI/ZarfixAICerdas1.0" \
    --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": "ZarfixAI/ZarfixAICerdas1.0",
		"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 "ZarfixAI/ZarfixAICerdas1.0" \
        --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": "ZarfixAI/ZarfixAICerdas1.0",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use ZarfixAI/ZarfixAICerdas1.0 with Docker Model Runner:
```
docker model run hf.co/ZarfixAI/ZarfixAICerdas1.0
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

ZarfixAI Cerdas 1.0

Summary:
ZarfixAI Cerdas 1.0 is a 4B parameter language model built for fast, efficient, and intelligent text generation.
Optimized for practical applications where cost, speed, and accuracy matter.

Based on janhq/Jan-v1-4B. Original work is licensed under Apache-2.0 (see LICENSE in this repo).

🚀 Features

4B parameters for a balance between performance and efficiency
Supports instruction-following and general conversation
Runs on consumer GPUs or cloud T4 instances for low-cost deployment
Apache-2.0 license — flexible for commercial and personal projects

🛠️ Quickstart

from transformers import AutoTokenizer, AutoModelForCausalLM

model_id = "ZarfixAI/ZarfixAICerdas1.0"

tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(model_id, device_map="auto")

prompt = "Explain the importance of renewable energy in simple terms."
inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
outputs = model.generate(**inputs, max_new_tokens=256)

print(tokenizer.decode(outputs[0], skip_special_tokens=True))

💡 Recommended Use Cases

Customer support bots
Knowledge assistants
Educational Q&A
Creative writing prompts
Lightweight RAG (Retrieval-Augmented Generation) systems

⚠️ Limitations

The model may produce inaccurate or biased outputs — always verify important information.
Not fine-tuned for high-risk applications (medical, legal, financial advice).

📜 License

Original model: janhq/Jan-v1-4B under Apache-2.0 license.
ZarfixAI Cerdas 1.0: Derivative work under the same Apache-2.0 license.
You are free to use, modify, and deploy, but must keep attribution to the original authors.

🙏 Acknowledgements

Special thanks to the developers of janhq/Jan-v1-4B for providing a strong open-source foundation.

Downloads last month: 8

Safetensors

Model size

4B params

Tensor type

BF16

Model tree for ZarfixAI/ZarfixAICerdas1.0

Base model

Qwen/Qwen3-4B-Thinking-2507

Finetuned

janhq/Jan-v1-4B

Finetuned

(5)

this model

Quantizations

2 models