Search Results for "tensor"

...The project recreates the core functionality of vLLM in a simplified architecture written in approximately a thousand lines of Python, making it easier for developers and researchers to understand how modern LLM inference systems work. Despite its compact design, nano-vllm incorporates advanced optimization techniques such as prefix caching, tensor parallelism, and CUDA graph execution to achieve high performance during model inference. The engine is intended primarily for educational use, experimentation, and lightweight deployments where a full production-grade inference stack may be unnecessary. Its API closely mirrors that of the original vLLM framework, allowing developers familiar with vLLM to adopt the tool with minimal changes.

Mooncake is an open-source infrastructure platform designed to optimize large language model serving by focusing on efficient management and transfer of model data and KV cache. The platform was originally developed as part of the serving infrastructure for the Kimi large language model system. Its architecture centers on a high-performance transfer engine that provides unified data transfer across different storage and networking technologies. This engine enables efficient movement of...

Megatron is a large, powerful transformer developed by the Applied Deep Learning Research team at NVIDIA. This repository is for ongoing research on training large transformer language models at scale. We developed efficient, model-parallel (tensor, sequence, and pipeline), and multi-node pre-training of transformer based models such as GPT, BERT, and T5 using mixed precision. Megatron is also used in NeMo Megatron, a framework to help enterprises overcome the challenges of building and training sophisticated natural language processing models with billions and trillions of parameters. ...

tiny-llm is an educational open-source project designed to teach system engineers how large language model inference and serving systems work by building them from scratch. The project is structured as a guided course that walks developers through the process of implementing the core components required to run a modern language model, including attention mechanisms, token generation, and optimization techniques. Rather than relying on high-level machine learning frameworks, the codebase uses...

...The project serves as a practical companion to a second edition of a PyTorch learning guide and is designed to help learners understand neural network concepts through hands-on coding examples. The repository covers a wide range of topics including tensor operations, neural network construction, model training workflows, and optimization strategies. It also introduces practical machine learning techniques such as convolutional neural networks, recurrent networks, and other architectures commonly used in modern AI applications. Each tutorial focuses on step-by-step implementation so learners can understand how theoretical concepts translate into working code. ...

...It has 456 billion total parameters with 45.9 billion activated per token and is trained with advanced parallel strategies such as LASP+, varlen ring attention, and Expert Tensor Parallelism, enabling a training context of 1 million tokens and up to 4 million tokens at inference. MiniMax-VL-01 extends this core by adding a 303M-parameter Vision Transformer and a two-layer MLP projector in a ViT–MLP–LLM framework, allowing the model to process images at dynamic resolutions up to 2016×2016.