Update README.md

README.md

@@ -1,187 +1,179 @@
 ---
 library_name: transformers
 license: other
-license_name: nvidia-internal-scientific-research-and-development-model-license
+license_name: nvidia-open-model-license
 license_link: >-
-  https://www.nvidia.com/en-us/agreements/enterprise-software/nvidia-internal-scientific-research-and-development-model-license/
+  https://www.nvidia.com/en-us/agreements/enterprise-software/nvidia-open-model-license/
 pipeline_tag: text-generation
 tags:
   - nvidia
   - pytorch
 ---
 
-# Nemotron-H-56B-Base-8K
+# OpenCodeReasoning-Distill-Qwen-7B-Instruct
 
 ## Model Overview 
 
-NVIDIA Nemotron-H-56B-Base-8K Base is a large language model (LLM) developed by NVIDIA that is designed as a completion model for a given piece of text. It uses a hybrid model architecture that consists primarily of Mamba-2 and MLP layers combined with just four Attention layers. The model features a context length of 8K. The supported languages include: English, German, Spanish, French, Italian, Korean, Portuguese, Russian, Japanese, and Chinese. For more detailed information on the model architecture, training, and evaluation, please see the [project page](https://research.nvidia.com/labs/adlr/nemotronh/) and the [technical report](https://arxiv.org/abs/2504.03624).
+Llama-3.1-Nemotron-Ultra-253B-v1 is a large language model (LLM) which is a derivative of [Meta Llama-3.1-405B-Instruct](https://huggingface.co/meta-llama/Llama-3.1-405B-Instruct) (AKA the *reference model*). It is a reasoning model that is post trained for reasoning, human chat preferences, and tasks, such as RAG and tool calling. The model supports a context length of 128K tokens. This model fits on a single 8xH100 node for inference.
 
-For best performance on a given task, users are encouraged to customize the model using the [NeMo Framework](https://docs.nvidia.com/nemo-framework/index.html) suite of customization tools including Parameter-Efficient Fine-Tuning (P-tuning, Adapters, LoRA, and more), and Model Alignment (SFT, SteerLM, RLHF, and more) using [NeMo-Aligner](https://github.com/NVIDIA/NeMo-Aligner).
+Llama-3.1-Nemotron-Ultra-253B-v1 is a model which offers a great tradeoff between model accuracy and efficiency. Efficiency (throughput) directly translates to savings. Using a novel Neural Architecture Search (NAS) approach, we greatly reduce the model’s memory footprint, enabling larger workloads, as well as reducing the number of GPUs required to run the model in a data center environment. This NAS approach enables the selection of a desired point in the accuracy-efficiency tradeoff. Furthermore, by using a novel method to vertically compress the model (see details [here](https://arxiv.org/abs/2503.18908)), it also offers a significant improvement in latency.
 
-This model is for research and development only.
+The model underwent a multi-phase post-training process to enhance both its reasoning and non-reasoning capabilities. This includes a supervised fine-tuning stage for Math, Code, Reasoning, Chat, and Tool Calling as well as multiple reinforcement learning (RL) stages using Group Relative Policy Optimization (GRPO) algorithms for reasoning, chat, and instruction-following. 
+
+This model is ready for commercial use. 
 
-This model is part of the Nemotron-H Collection. You can find the models in this family here:
-- [Nemotron-H-56B-Base-8K](https://huggingface.co/nvidia/Nemotron-H-56B-Base-8K)
-- [Nemotron-H-47B-Base-8K](https://huggingface.co/nvidia/Nemotron-H-47B-Base-8K)
-- [Nemotron-H-8B-Base-8K](https://huggingface.co/nvidia/Nemotron-H-8B-Base-8K)
 
 ## License/Terms of Use
 
-GOVERNING TERMS: Use of this model is governed by the [NVIDIA Internal Scientific Research and Development Model License](https://www.nvidia.com/en-us/agreements/enterprise-software/nvidia-internal-scientific-research-and-development-model-license/).
+GOVERNING TERMS: Your use of this model is governed by the [NVIDIA Open Model License.](https://www.nvidia.com/en-us/agreements/enterprise-software/nvidia-open-model-license/) Additional Information: [Llama 3.1 Community License Agreement](https://www.llama.com/llama3\_1/license/). Built with Qwen.
 
 **Model Developer:** NVIDIA
 
-**Model Dates:**
-
-October 2024 - March 2025
-
-**Data Freshness:**
-
-September 2024
-
-The pretraining data has a cutoff date of September 2024. 
-
+**Model Dates:** Trained between February 2025 and March 2025
 
-## Use Case:
 
-This model is intended for developers and researchers building LLMs.
+### Use Case:
+Developers designing AI-powered code generation applications. Also suitable for typical instruction-following tasks. 
 
 ## Release Date: 
 
-4/14/2025
+2025-04-21
 
 ## References
 
-- [\[2504.03624\] Nemotron-H: A Family of Accurate and Efficient Hybrid Mamba-Transformer Models](https://arxiv.org/abs/2504.03624)
+- [\[2504.01943\] OpenCodeReasoning: Advancing Data Distillation for Competitive Coding](https://arxiv.org/abs/2504.01943)
 
 ## Model Architecture
-- Architecture Type: Hybrid Mamba-Transformer
-- Network Architecture: Nemotron-H
+- Architecture Type: Dense decoder-only Transformer model
+- Network Architecture: Qwen
 
-This model has 56B of model parameters.
+**This model was developed based on Qwen2.5-7B-Instruct <br>
+** This model has 7B of model parameters. <br>
 
-## Input
-- Input Type(s): Text 
-- Input Format(s): String
-- Input Parameters: One-Dimensional (1D): Sequences
-- Other Properties Related to Input: Context length up to 8K. Supported languages include German, Spanish, French, Italian, Korean, Portuguese, Russian, Japanese, Chinese and English.
+## Intended use
 
-## Output
-- Output Type(s): Text 
-- Output Format: String
-- Output Parameters: One-Dimensional (1D): Sequences
+OpenCodeReasoning-Distill-Qwen-7B-Instruct is a competitive code generation focused reasoning and chat model intended to be used in English. 
 
-Our AI models are designed and/or optimized to run on NVIDIA GPU-accelerated systems. By leveraging NVIDIA’s hardware (e.g. GPU cores) and software frameworks (e.g., CUDA libraries), the model achieves faster training and inference times compared to CPU-only solutions. <br> 
+## Input  
+- **Input Type:** Text  
+- **Input Format:** String  
+- **Input Parameters:** One-Dimensional (1D)  
+- **Other Properties Related to Input:** Context length up to 32,768 tokens
 
-## Software Integration
-- Runtime Engine(s): NeMo 24.12
-- Supported Hardware Microarchitecture Compatibility: NVIDIA H100-80GB, NVIDIA A100
-- Operating System(s): Linux
+## Output  
+- **Output Type:** Text  
+- **Output Format:** String  
+- **Output Parameters:** One-Dimensional (1D)  
+- **Other Properties Related to Output:** Context length up to 32,768 tokens
 
-## Model Version
-- v1.0
+## Software Integration  
+- **Runtime Engine:** Transformers  
+- **Recommended Hardware Microarchitecture Compatibility:**   
+  - NVIDIA Hopper  
+  - NVIDIA Ampere  
+- **Preferred Operating System(s):** Linux 
 
-## Prompt Format
+## Model Version  
+1.0 (4/21/2025)
 
-As this is a base model, no explicit prompt format is recommended or required. 
+## Quick Start and Usage Recommendations:
 
-### Example
+We recommend setting temperature to \`0.6\`, and Top P to \`0.95\` for inference on LiveCodeBench.
 
-```python
-import torch
-from transformers import AutoTokenizer, AutoModelForCausalLM
-
-# Load the tokenizer and model
-tokenizer  = AutoTokenizer.from_pretrained("nvidia/Nemotron-H-56B-Base-8K", trust_remote_code=True)
-model = AutoModelForCausalLM.from_pretrained("nvidia/Nemotron-H-56B-Base-8K", torch_dtype=torch.bfloat16, trust_remote_code=True, device_map="auto")
+### Use It with Transformers
+See the snippet below for usage with [Hugging Face Transformers](https://huggingface.co/docs/transformers/main/en/index) library. Please see the example below.
 
-prompt = "When was NVIDIA founded?"
+We recommend using the *transformers* package with version 4.48.3.   
+Example:
 
-outputs = model.generate(**tokenizer(prompt, return_tensors="pt", add_special_tokens=False).to(model.device))
-print(tokenizer.decode(outputs[0]))
+```py
+import torch
+import transformers
+
+model_id = "nvidia/OpenCodeReasoning-Distill-Qwen-7B-Instruct"
+model_kwargs = {"torch_dtype": torch.bfloat16, "trust_remote_code": True, "device_map": "auto"}
+tokenizer = transformers.AutoTokenizer.from_pretrained(model_id)
+tokenizer.pad_token_id = tokenizer.eos_token_id
+
+pipeline = transformers.pipeline(
+   "text-generation",
+   model=model_id,
+   tokenizer=tokenizer,
+   max_new_tokens=32768,
+   temperature=0.6,
+   top_p=0.95,
+   **model_kwargs
+)
+
+print(pipeline([{"role": "user", "content": "Solve x*(sin(x)+2)=0"}]))
 ```
 
-## Training, Testing, and Evaluation Datasets
+## Training and Evaluation Datasets
 
-### Training & Testing Datasets:
+## Training Datasets
 
-The training corpus for Nemotron-H-56B-Base-8K Base consists of English and multilingual text (German, Spanish, French, Italian, Korean, Portuguese, Russian, Japanese, Chinese and English), as well as code. Our sources cover a variety of document types such as: webpages, dialogue, articles, and other written materials. This model was also improved using synthetic data from Qwen (Built with Qwen). The corpus spans domains including legal, math, science, finance, and more. We also include a small portion of question-answering, and alignment style data to improve model accuracies.
+This model is trained using [OpenCodeReasoning](https://huggingface.co/datasets/nvidia/OpenCodeReasoning) dataset.
 
-**Data Collection for Training & Testing Datasets:**
-Hybrid: Automated, Human, Synthetic
+**Data Collection for Training Datasets:**
 
-**Data Labeling for Training & Testing Datasets:**
-Hybrid: Automated, Human, Synthetic
+- Hybrid: Automated, Human, Synthetic
 
-### Evaluation Datasets 
+**Data Labeling for Training Datasets:**
 
-We used the datasets listed in the next section to evaluate Nemotron-H-56B-Base-8K Base. 
+- Hybrid: Automated, Human, Synthetic
 
-Data Collection for Evaluation Datasets:
-Hybrid: Human, Synthetic
+## Evaluation Datasets 
 
-Data Labeling for Evaluation Datasets:
-Hybrid: Human, Synthetic, Automatic
-
-#### Commonsense Understanding Evaluations:
-
-| ARC Challenge 25-shot | Hellaswag 10-shot | Winogrande 5-shot | CommonsenseQA 7-shot |
-|-------------|--------------|-----------------|------------------|
-| 94.97 | 89.00 | 84.45 | 86.73 |
+We used the datasets listed in the next section to evaluate Llama-3.1-Nemotron-Ultra-253B-v1. 
 
-- ARC (Ai2 reasoning challenge)-Challenge - The challenge set of questions from a benchmark that contains grade-school level, multiple-choice science questions to assess question answering ability of language models. [Dataset](https://huggingface.co/datasets/allenai/ai2_arc)
-- Hellaswag - Tests the ability of a language model to correctly finish the provided context from a choice of possible options. [Dataset](https://huggingface.co/datasets/Rowan/hellaswag )
-- Winogrande - Tests the ability to choose the right option for a given sentence which requires commonsense reasoning. [Dataset](https://huggingface.co/datasets/allenai/winogrande )
-- CommonsenseQA - A multiple-choice question answering dataset that requires different types of commonsense knowledge to predict the correct answers. [Dataset](https://huggingface.co/datasets/tau/commonsense_qa  )
+Data Collection for Evaluation Datasets:
 
-#### Coding Evaluations:
+- Hybrid: Human/Synthetic
 
-| MBPP (sanitized) 3-shot | MBPP+ 0-shot | HumanEval 0-shot | HumanEval+ 0-shot |
-|-------------|--------------|-----------------|------------------|
-| 77.82 | 67.20| 60.37 | 54.27 |
+Data Labeling for Evaluation Datasets:
 
-- MBPP (Mostly Basic Python Programming Problems) - Evaluates ability to generate solutions for Python programming tasks. [Dataset](https://github.com/google-research/google-research/tree/master/mbpp)
-- MBPP+ - Extended version of MBPP with additional validation. [Dataset](https://huggingface.co/datasets/evalplus/mbppplus)
-- HumanEval - Tests code generation and completion abilities in Python. [Dataset](https://github.com/openai/human-eval)
+- Hybrid: Human/Synthetic/Automatic
 
-#### Math Evaluations:
 
+## Evaluation Results
 
-| GSM8K 8-shot CoT |  MATH 4-shot CoT |  MATH-Lvl 5 4-shot CoT  | MATH-500 4-shot CoT |
-|--------------|------------|------------|------------|
-| 93.71 | 59.42 | 35.19 | 57.37 |
+### LiveCodeBench (20240801-20250201)
 
-- GSM8K (Grade School Math 8K) - Evaluates grade school level mathematical word problem solving. [Dataset](https://github.com/openai/grade-school-math)
-- MATH - Tests mathematical ability across multiple difficulty levels and various subjects including: Prealgebra, Algebra, Number Theory, Counting and Probability, Geometry, Intermediate Algebra, and Precalculus. [Dataset](https://github.com/hendrycks/math)
-- MATH Lvl 5 - Only the most difficult questions from the MATH dataset. [Dataset](https://github.com/hendrycks/math)
-- MATH-500 - Tests advanced mathematical problem solving across algebra, geometry, and calculus. [Dataset](https://huggingface.co/datasets/HuggingFaceH4/MATH-500)
+| Models | Pass@1 |
+|--------------|------------|
+| R1-Distill-Qwen-7B | 37.6 |
+| OpenCodeReasoning-Distill-Qwen-7B-Instruct | 51.3 | 
 
+User Prompt Template (without starter code):
 
-#### General Evaluations:
+````
+"You will be given a question (problem specification) and will generate a correct Python program that matches the specification and passes all tests.
 
+Question: {prompt}
 
-| MMLU-Pro 5-shot CoT | MMLU 5-shot|
-|-------------------|------------------|
-|60.51 |84.21 | 
+Read the inputs from stdin solve the problem and write the answer to stdout (do not directly test on the sample inputs). Enclose your code within delimiters as follows. Ensure that when the python program runs, it reads the inputs, runs the algorithm and writes output to STDOUT.
+```python
+# YOUR CODE HERE
+```
+````
 
-- MMLU Pro - Evaluates language understanding models across a broad range of challenging, reasoning-focused questions across 14 diverse domains.
-[Dataset](https://huggingface.co/datasets/TIGER-Lab/MMLU-Pro)
-- MMLU - Tests knowledge across 57 subjects including science, humanities, math and more. [Dataset](https://github.com/hendrycks/test)
+User Prompt Template (with starter code):
 
-## Potential Known Risks for Usage
+````
+You will be given a question (problem specification) and will generate a correct Python program that matches the specification and passes all tests.
 
-The model was trained on data that contains toxic language, unsafe content, and societal biases originally crawled from the internet. Therefore, the model may amplify those biases and return toxic responses especially when prompted with toxic prompts. The model may generate answers that may be inaccurate, omit key information, or include irrelevant or redundant text producing socially unacceptable or undesirable text, even if the prompt itself does not include anything explicitly offensive.
+Question: {prompt}
 
-The model demonstrates weakness to indirect prompt injection via some encodings, including Base16, Hex/ASCII, and Braille, though is more resilient than other similar models to injections using the more common Base64 vector.
+You will use the following starter code to write the solution to the problem and enclose your code within delimiters.
+```python
+{starter_code}
+```
+````
 
-## Inference
-- Engine: NeMo
-- Test Hardware NVIDIA H100-80GB
+## Ethical Considerations:
 
-## Ethical Considerations
-NVIDIA believes Trustworthy AI is a shared responsibility and we have established policies and practices to enable development for a wide array of AI applications. When downloaded or used in accordance with our terms of service, developers should work with their internal model team to ensure this model meets requirements for the relevant industry and use case and addresses unforeseen product misuse. 
+NVIDIA believes Trustworthy AI is a shared responsibility and we have established policies and practices to enable development for a wide array of AI applications.  When downloaded or used in accordance with our terms of service, developers should work with their internal model team to ensure this model meets requirements for the relevant industry and use case and addresses unforeseen product misuse. 
 
-For more detailed information on ethical considerations for this model, please see the Responsible Use Guide available at http://nvidia.com/nemotron-responsible-use.
+For more detailed information on ethical considerations for this model, please see the Model Card++ [Explainability](./EXPLAINABILITY.md), [Bias](./BIAS.md), [Safety & Security](./SAFETY_and_SECURITY.md), and [Privacy](./PRIVACY.md) Subcards.
 
 Please report security vulnerabilities or NVIDIA AI Concerns [here](https://www.nvidia.com/en-us/support/submit-security-vulnerability/).
-