hadadrjt/JARVIS · Hugging Face

J.A.R.V.I.S., which stands for "Just a Rather Very Intelligent System", is an advanced AI language model inspired by Iron Man iconic assistant. This model delivers context-aware, high-fidelity natural language processing capabilities for a wide range of applications.

NOTICE! This is only the base model and a lighter version for public release. To use the more advanced version, please use the available space provided.

Model Highlights

Powerful Text Generation Engine
J.A.R.V.I.S. serves as the core of an AI solution capable of producing natural and creative text with high quality, ready to help you create inspiring and engaging content.
Seamless Integration, Limitless Possibilities
Designed with high flexibility, J.A.R.V.I.S. can be seamlessly integrated into a wide range of platforms and systems you use, including web applications, chatbots, agent modes, tools, system integrations, and automated workflows, adapting effortlessly to available resource capacities.
Scalable & Resource-Aware
Regardless of your computing environment, J.A.R.V.I.S. can be optimized for optimal performance, from lightweight devices to enterprise-class infrastructure, ensuring efficiency and responsiveness without compromise.
Versatile Use Cases
Ideal for a wide range of needs, from creative content creation, writing assistance, to developing interactive text-based applications, giving you unlimited freedom to innovate.
Agentic Use
Example usage, please refer to the Qwen3 Documentation, and the Qwen-Agent GitHub repository.

J.A.R.V.I.S. is not just an AI model, it is an intelligent partner ready to transform your ideas into words with a touch of intelligence and high flexibility.

For more detailed information about this model, you can explore all its advantages by referring to the explanation of the underlying base model. Here!

Core Capabilities

Discover the amazing power of J.A.R.V.I.S., as your smart and reliable AI companion. It answers your toughest questions with confidence, explains complex ideas in simple terms, translates languages smoothly, writes clean and efficient code, summarizes lengthy information into easy-to-understand points, sparks your creativity with fresh and original content, and guides you through study or research with expert advice. This is only the beginning of what J.A.R.V.I.S. can do for you.

Notebook

Although this model has a large size and many parameters, you can run it on Google Colab for free! Using the script I have optimized. Click here to access it.

Ollama

The source of the J.A.R.V.I.S. model on Ollama can be found here.

ollama run hadad/JARVIS

# 4-bit (Q4_K_M)
ollama run hadad/JARVIS:4bit

Quick Start

import torch
from transformers import AutoModelForCausalLM, AutoTokenizer, TextStreamer

# Model
model = "hadadrjt/JARVIS"

# Load tokenizer and model
tokenizer = AutoTokenizer.from_pretrained(model)
model = AutoModelForCausalLM.from_pretrained(
    model,
    torch_dtype="auto",
    device_map="auto"
)

# User input
input = "Tell me about yourself" # Insert your message here.

# Build chat-style prompt
messages = [{"role": "user", "content": input}]

text = tokenizer.apply_chat_template(
    messages,
    tokenize=False,
    add_generation_prompt=True,
    enable_thinking=True
)

# Tokenize input
model_inputs = tokenizer([text], return_tensors="pt").to(model.device)

# Streaming setup
streamer = TextStreamer(tokenizer, skip_prompt=True, skip_special_tokens=True)

# Generate response with streaming
with torch.inference_mode():
    generated_ids = model.generate(
        **model_inputs,
        max_new_tokens=32768,
        temperature=0.6,
        top_p=0.95,
        min_p=0,
        top_k=20,
        repetition_penalty=1.0,
        streamer=streamer
    )

Responsible Use

Although J.A.R.V.I.S. delivers impressive results, it may reflect biases from its training data and occasionally produce incorrect outputs. It is not intended for real-time safety-critical applications without human oversight. I recommend implementing human review workflows, monitoring outputs for fairness and accuracy, and updating the model with domain-specific data over time.

Acknowledgments and Contact

This work builds upon the Qwen3-14B open-source foundation. For feedback, collaboration or support, please reach out to Hadad Darajat at [email protected]. I welcome contributions that help J.A.R.V.I.S. continue to evolve.

Thank you for choosing J.A.R.V.I.S. to power your intelligent language applications.

Citation

@misc{jarvis,
  title   = {J.A.R.V.I.S.},
  author  = {Hadad Darajat},
  year    = {2025},
  license = {Apache-2.0},
  base    = {Qwen/Qwen3-14B}
}

hadadrjt
/

JARVIS