You like pytorch? You like micrograd? You love tinygrad! ❤️

SGLang is a fast serving framework for large language models and vision language models.

A unified inference and post-training framework for accelerated video generation.

A high-throughput and memory-efficient inference and serving engine for LLMs

tiny corporation website

Mirage Persistent Kernel: Compiling LLMs into a MegaKernel

Home for "How To Scale Your Model", a short blog-style textbook about scaling LLMs on TPUs

[ACL 2025 main] FR-Spec: Frequency-Ranked Speculative Sampling

Deep learning at the speed of light.

Structured Text Generation

A minimal PyTorch implementation of probabilistic diffusion models for 2D datasets.

Train to 94% on CIFAR-10 in <6.3 seconds on a single A100. Or ~95.79% in ~110 seconds (or less!)

Official inference repo for FLUX.1 models

HunyuanVideo: A Systematic Framework For Large Video Generation Model

Train transformer language models with reinforcement learning.

Official repository of ’Visual-RFT: Visual Reinforcement Fine-Tuning’

Finetune Llama 3.3, DeepSeek-R1 & Reasoning LLMs 2x faster with 70% less memory! 🦥