| --- |
| license: apache-2.0 |
| datasets: |
| - nvidia/OpenScienceReasoning-2 |
| language: |
| - en |
| base_model: |
| - Qwen/Qwen3-1.7B |
| pipeline_tag: text-generation |
| library_name: transformers |
| tags: |
| - trl |
| - text-generation-inference |
| - medical |
| - science |
| --- |
| |
|  |
|
|
| # **OpenScienceReasoning-Qwen-e10** |
|
|
| > OpenScienceReasoning-Qwen-e10 is a high-efficiency, science-focused reasoning model fine-tuned on **Qwen3-1.7B** using the [**nvidia/OpenScienceReasoning-2**](https://huggingface.co/datasets/nvidia/OpenScienceReasoning-2) dataset. It incorporates **10,000 distinct entries** for scientific reasoning, chain-of-thought exploration, and analytical problem solving. |
| > The model blends symbolic precision, scientific logic, and structured output fluency—making it an ideal tool for researchers, educators, and developers seeking advanced reasoning under constrained compute. |
|
|
| > \[!note] |
| > GGUF: [https://huggingface.co/prithivMLmods/OpenScienceReasoning-Qwen-e10-GGUF](https://huggingface.co/prithivMLmods/OpenScienceReasoning-Qwen-e10-GGUF) |
|
|
| --- |
|
|
| ## **Key Features** |
|
|
| 1. **Scientific Reasoning & Chain-of-Thought** |
| Fine-tuned on **10,000 curated entries** from the **OpenScienceReasoning-2** dataset, designed to enhance step-by-step analytical reasoning in science and mathematics. |
|
|
| 2. **Advanced Code Reasoning & Generation** |
| Supports multi-language coding with explanations, optimization hints, and error detection—ideal for algorithm synthesis, debugging, and prototyping. |
|
|
| 3. **Mathematical & Scientific Problem Solving** |
| Performs analytical reasoning in physics, biology, chemistry, and mathematics—explaining concepts, solving equations, and handling symbolic derivations. |
|
|
| 4. **Hybrid Symbolic-AI Thinking** |
| Combines structured logic, chain-of-thought reasoning, and open-ended inference, delivering robust performance on STEM-related tasks. |
|
|
| 5. **Structured Output Mastery** |
| Seamlessly generates output in **LaTeX**, **Markdown**, **JSON**, **CSV**, and **YAML**, suited for technical documentation, research papers, and structured data. |
|
|
| 6. **Optimized Lightweight Footprint for Versatile Deployment** |
| Balances performance and efficiency, making it deployable on **mid-range GPUs**, **offline clusters**, and **edge AI systems**. |
|
|
| --- |
|
|
| ## **Quickstart with Transformers** |
|
|
| ```python |
| from transformers import AutoModelForCausalLM, AutoTokenizer |
| |
| model_name = "prithivMLmods/OpenScienceReasoning-Qwen-e10" |
| |
| model = AutoModelForCausalLM.from_pretrained( |
| model_name, |
| torch_dtype="auto", |
| device_map="auto" |
| ) |
| tokenizer = AutoTokenizer.from_pretrained(model_name) |
| |
| prompt = "Explain the difference between Newtonian mechanics and quantum mechanics with examples." |
| |
| messages = [ |
| {"role": "system", "content": "You are a scientific tutor skilled in reasoning, math, and coding."}, |
| {"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] |
| print(response) |
| ``` |
|
|
| --- |
|
|
| ## **Intended Use** |
|
|
| * Scientific tutoring, computational reasoning, and mathematical education |
| * Research assistant for physics, chemistry, biology, and interdisciplinary domains |
| * Structured technical data generation in multiple formats |
| * STEM-focused chatbot or API for research and education tools |
| * Deployment in mid-resource environments requiring high reasoning fidelity |
|
|
| ## **Limitations** |
|
|
| * Not tuned for general-purpose or long-form creative writing |
| * Context limitations may hinder multi-document or full codebase analysis |
| * Specialized for scientific and technical reasoning—general chat may underperform |
| * Prioritizes structured logic over casual or emotional tone generation |
