Delta Pavonis Qwen 14B

🚀 Delta-Pavonis-Qwen-14B

Delta-Pavonis-Qwen-14B is based on the Qwen 2.5 14B modality architecture, aiming to enhance the reasoning capabilities of 14B-parameter models. It excels in general - purpose reasoning, answering, and multi - step problem - solving.

✨ Features

Key Improvements

1. Enhanced General Knowledge: The model offers extensive knowledge across diverse domains, enabling more accurate question - answering and coherent response generation.
2. Improved Instruction Following: It has made significant progress in understanding and executing complex instructions, generating structured responses, and maintaining coherence during long - term interactions.
3. Versatile Adaptability: It can better handle various prompts, enhancing its ability to deal with a wide range of topics and conversation styles, including open - ended and structured inquiries.
4. Long - Context Support: It supports up to 128K tokens for input context and can generate up to 8K tokens in a single output, perfect for detailed responses.

🚀 Quick Start

Basic Usage

Here is a code snippet with apply_chat_template to show you how to load the tokenizer and model and generate content:

from transformers import AutoModelForCausalLM, AutoTokenizer

model_name = "prithivMLmods/Delta-Pavonis-Qwen-14B"

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

prompt = "What are the key principles of general-purpose AI?"
messages = [
    {"role": "system", "content": "You are a helpful assistant capable of answering a wide range of questions."},
    {"role": "user", "content": prompt}
]
text = tokenizer.apply_chat_template(
    messages,
    tokenize=False,
    add_generation_prompt=True
)
model_inputs = tokenizer([text], return_tensors="pt").to(model.device)

generated_ids = model.generate(
    **model_inputs,
    max_new_tokens=512
)
generated_ids = [
    output_ids[len(input_ids):] for input_ids, output_ids in zip(model_inputs.input_ids, generated_ids)
]

response = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)[0]

📚 Documentation

Intended Use

1. General - Purpose Reasoning: Designed for wide - ranging applications, it can assist with logical reasoning, answer diverse questions, and solve general knowledge problems.
2. Educational and Informational Assistance: Suitable for providing explanations, summaries, and research - based responses for students, educators, and general users.
3. Conversational AI and Chatbots: Ideal for building intelligent conversational agents that require contextual understanding and dynamic response generation.
4. Multilingual Applications: Supports global communication, translations, and multilingual content generation.
5. Structured Data Processing: Capable of analyzing and generating structured outputs, such as tables and JSON, useful for data science and automation.
6. Long - Form Content Generation: Can generate extended responses, including articles, reports, and guides, while maintaining coherence in large text outputs.

Limitations

1. Hardware Requirements: Due to its large parameter size and long - context support, it requires high - memory GPUs or TPUs.
2. Potential Bias in Responses: Although designed to be neutral, outputs may still reflect biases in the training data.
3. Inconsistent Outputs in Creative Tasks: May produce variable results in storytelling and highly subjective topics.
4. Limited Real - World Awareness: It has no access to real - time events beyond its training cutoff.
5. Error Propagation in Extended Outputs: Minor errors in early responses may affect the overall coherence of long - form outputs.
6. Prompt Sensitivity: The effectiveness of responses may depend on the quality of the input prompt structure.

📄 License

The model is licensed under the Apache - 2.0 license.