Nemotron 3.5 Transcribe Integrations

​Transformers is a library of pretrained natural language processing, computer vision, audio, and multimodal models for inference and training. Use Transformers to train models on your data, build inference applications, and generate text with large language models. Explore the Hugging Face Hub today to find a model and use Transformers to help you get started right away.​ Simple and optimized inference class for many machine learning tasks like text generation, image segmentation, automatic speech recognition, document question answering, and more. A comprehensive trainer that supports features such as mixed precision, torch.compile, and FlashAttention for training and distributed training for PyTorch models.​ Fast text generation with large language models and vision language models. Every model is implemented from only three main classes (configuration, model, and preprocessor) and can be quickly used for inference or training.

Explore the latest optimized AI models, connect AI agents to data with NVIDIA NeMo, and deploy anywhere with NVIDIA NIM microservices. NVIDIA NIM is a set of easy-to-use inference microservices that facilitate the deployment of foundation models across any cloud or data center, ensuring data security and streamlined AI integration. Additionally, NVIDIA AI provides access to the Deep Learning Institute (DLI), offering technical training to gain in-demand skills, hands-on experience, and expert knowledge in AI, data science, and accelerated computing. AI models generate responses and outputs based on complex algorithms and machine learning techniques, and those responses or outputs may be inaccurate, harmful, biased, or indecent. By testing this model, you assume the risk of any harm caused by any response or output of the model. Please do not upload any confidential information or personal data unless expressly permitted. Your use is logged for security purposes.

vLLM is a high-performance library designed to facilitate efficient inference and serving of Large Language Models (LLMs). Originally developed in the Sky Computing Lab at UC Berkeley, vLLM has evolved into a community-driven project with contributions from both academia and industry. It offers state-of-the-art serving throughput by efficiently managing attention key and value memory through its PagedAttention mechanism. It supports continuous batching of incoming requests and utilizes optimized CUDA kernels, including integration with FlashAttention and FlashInfer, to enhance model execution speed. Additionally, vLLM provides quantization support for GPTQ, AWQ, INT4, INT8, and FP8, as well as speculative decoding capabilities. Users benefit from seamless integration with popular Hugging Face models, support for various decoding algorithms such as parallel sampling and beam search, and compatibility with NVIDIA GPUs, AMD CPUs and GPUs, Intel CPUs, and more.