Instructions to use Sergey321-345/XENON-AI-FULL with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Sergey321-345/XENON-AI-FULL with Transformers:

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

pipe = pipeline("text-generation", model="Sergey321-345/XENON-AI-FULL")
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("Sergey321-345/XENON-AI-FULL")
model = AutoModelForCausalLM.from_pretrained("Sergey321-345/XENON-AI-FULL", 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 Sergey321-345/XENON-AI-FULL with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Sergey321-345/XENON-AI-FULL"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Sergey321-345/XENON-AI-FULL",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Sergey321-345/XENON-AI-FULL

SGLang

How to use Sergey321-345/XENON-AI-FULL 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 "Sergey321-345/XENON-AI-FULL" \
    --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": "Sergey321-345/XENON-AI-FULL",
		"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 "Sergey321-345/XENON-AI-FULL" \
        --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": "Sergey321-345/XENON-AI-FULL",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Sergey321-345/XENON-AI-FULL with Docker Model Runner:
```
docker model run hf.co/Sergey321-345/XENON-AI-FULL
```

XENON Quanta 1

A 4B Language Model by XenLogic

XENON Quanta 1 is a fine-tuned language model developed by XenLogic, based on Qwen3-4B-Instruct-2507.

Designed for conversational assistance, mathematical problem-solving, educational explanations, and programming support.

Model Overview

Property	Description
Model	XENON Quanta 1
Developer	XenLogic
Base model	Qwen3-4B-Instruct-2507
Parameters	~4 billion
Architecture	Transformer
Fine-tuning	LoRA, merged weights
Weight format	Safetensors
Current stage	Early research release

Capabilities

Natural-language conversations
Mathematics and structured problem-solving
Educational assistance
Python and general programming support
Experimental reasoning-oriented responses

Evaluation

XENON Quanta 1 has undergone preliminary manual evaluation using mathematical word problems.

The evaluation identified both successful solutions and limitations involving arithmetic accuracy, output formatting, and response completion.

These early tests are not a standardized benchmark.

Research and Documentation

View evaluation reports

Limitations

The model may produce incorrect calculations, incomplete responses, or formatting artifacts. Important results should be independently verified.

Developed by XenLogic.

Downloads last month: 1,419

Safetensors

Model size

4B params

Tensor type

F16

Model tree for Sergey321-345/XENON-AI-FULL

Base model

Qwen/Qwen3-4B-Instruct-2507

Finetuned

(2,748)

this model

Sergey321-345
/

XENON-AI-FULL