From a0abe1c8720b591c8499acd8a9ec00c41209a1b1 Mon Sep 17 00:00:00 2001 From: Marc Romeyn Date: Thu, 18 Dec 2025 16:31:51 +0100 Subject: [PATCH] Adding video Signed-off-by: Marc Romeyn --- README.md | 33 +++++++++++++++++++++++++++++---- docs/index.md | 15 ++++++++++++++- 2 files changed, 43 insertions(+), 5 deletions(-) diff --git a/README.md b/README.md index aaa802653..a958f49d7 100644 --- a/README.md +++ b/README.md @@ -1,12 +1,31 @@ # NVIDIA Nemotron Developer Repository -Developer companion repo for working with NVIDIA's Nemotron models: inference, fine-tuning, agents, visual reasoning, deployment, and complete training recipes. +**Open and efficient models for agentic AI** — training recipes, deployment guides, and use-case examples for the Nemotron family. [![Python 3.10+](https://img.shields.io/badge/python-3.10%2B-blue.svg)](https://www.python.org/downloads/) [![License: Apache 2.0](https://img.shields.io/badge/License-Apache%202.0-green.svg)](https://opensource.org/licenses/Apache-2.0) [![Contributions Welcome](https://img.shields.io/badge/contributions-welcome-brightgreen.svg)](CONTRIBUTING.md) [![Docs](https://img.shields.io/badge/docs-dev-76B900.svg)](https://nvidia-nemo.github.io/Nemotron/dev/) +
+ +[![Watch the Nemotron Overview](https://img.youtube.com/vi/_y9SEtn1lU8/maxresdefault.jpg)](https://www.youtube.com/watch?v=_y9SEtn1lU8) + +**[Watch: Nemotron Overview](https://www.youtube.com/watch?v=_y9SEtn1lU8)** + +
+ +--- + +## Why Nemotron? + +| | | +|---|---| +| **Open Models** | Fully transparent training data, techniques, and weights for community innovation | +| **Compute Efficiency** | Model pruning and optimization enabling higher throughput via TensorRT-LLM | +| **High Accuracy** | Built on frontier open models with human-aligned reasoning for agentic workflows | +| **Flexible Deployment** | Deploy anywhere — edge, single GPU, or data center with NIM microservices | + --- ## Repository Overview @@ -25,9 +44,15 @@ nemotron/ ## What is Nemotron? -[NVIDIA Nemotron](https://developer.nvidia.com/nemotron) is a family of open, high-efficiency models with fully transparent training data, weights, and recipes. +[NVIDIA Nemotron](https://www.nvidia.com/en-us/ai-data-science/foundation-models/nemotron/) is a family of open, high-efficiency multimodal models purpose-built for agentic AI. + +**Model Tiers:** + +- **Nano** — Optimized for edge and PC deployments +- **Super** — Single GPU deployment with highest throughput +- **Ultra** — Multi-GPU datacenter applications -Nemotron models are designed for **agentic AI workflows**—they excel at coding, math, scientific reasoning, tool calling, instruction following, and visual reasoning. Models are optimized for deployment across edge, single GPU, and data center environments, with support for NeMo, TensorRT-LLM, vLLM, SGLang, and NIM microservices. +Nemotron models excel at coding, math, scientific reasoning, tool calling, instruction following, and visual reasoning. Deploy across edge, single GPU, or data center environments with support for NeMo, TensorRT-LLM, vLLM, SGLang, and NIM microservices. --- @@ -136,4 +161,4 @@ Apache 2.0 License — see [LICENSE](LICENSE) for details. --- -**NVIDIA Nemotron** — Open, transparent, and reproducible. +**NVIDIA Nemotron** — Open and efficient models for agentic AI. diff --git a/docs/index.md b/docs/index.md index d6152a8c2..72a8b446e 100644 --- a/docs/index.md +++ b/docs/index.md @@ -1,6 +1,10 @@ # Nemotron Training Recipes -Reproducible training recipes for the NVIDIA Nemotron model family — transparent pipelines for data preparation, training, and evaluation across all stages. +**Open and efficient models for agentic AI** — reproducible training pipelines with fully transparent data, techniques, and weights. + +
+ +
## Quick Start @@ -71,6 +75,15 @@ The Nemotron training pipeline follows a three-stage approach with full artifact | 1 | [SFT](train/nano3/sft.md) | Supervised fine-tuning for instruction following | | 2 | [RL](train/nano3/rl.md) | Reinforcement learning for alignment | +## Why Nemotron? + +| | | +|---|---| +| **Open Models** | Transparent training data, techniques, and weights for community innovation | +| **Compute Efficiency** | Model pruning enabling higher throughput via TensorRT-LLM | +| **High Accuracy** | Built on frontier open models with human-aligned reasoning | +| **Flexible Deployment** | Deploy anywhere — edge, single GPU, or data center with NIM | + ## Key Features - **Complete Pipelines** — From raw data to deployment-ready models