AI Train2020
PyTorch library for operator learning: neural networks that map between whole function spaces, not fixed grids, so a model trained at one resolution runs at any other. Bundles FNO, Tensorized FNO and related architectures, mainly for solving PDEs.
AI Infra2020
Sits between PyTorch and micrograd: eager tensors with autograd plus a small, fully hackable compiler that fuses operations into kernels. Adding a new accelerator backend takes about 25 low-level ops, so it runs on CUDA, Metal, AMD, and WebGPU.
AI Infra2021
Runs, manages, and scales AI workloads across 20+ clouds, Kubernetes, Slurm, and on-prem from one YAML or Python spec. Auto-provisions GPUs/TPUs, fails over across regions and providers when capacity is short, and routes jobs to the cheapest option.
AI Train2021
HPC-AI Technology Inc. (Colossal-AI team), Shenggui Li +1HPC-AI Technology Inc., National University of Singapore
Scales a single-GPU training script to thousands of GPUs through a unified interface, combining data, pipeline, tensor, and sequence parallelism. Its Gemini memory manager offloads tensors across GPU, CPU, and NVMe so models far larger than VRAM still fit.
MLOps2021
Orchestrates ML training pipelines and production agent workflows from one Python codebase that runs unchanged from a laptop to Kubernetes or any cloud. Auto-versions artifacts, models, and agent checkpoints, with no orchestrator or framework lock-in.
AI Infra2022
Fused CUDA kernels that compute exact attention without ever writing the full N×N score matrix to GPU memory, cutting memory from quadratic to linear and speeding up training and inference on A100/H100. Ships FlashAttention-2/3 plus KV-cache decode paths.
AI Image2022
Unifies successive YOLO generations — YOLOv8, YOLO11, YOLOv3 and newer — under one package and a single `YOLO` API spanning detection, segmentation, classification, pose, oriented boxes and tracking, plus one-line export to ONNX, TensorRT and CoreML.
AI Train2022
Reproduces GPT-2 (124M) from scratch on OpenWebText in ~4 days on an 8xA100 node, with the whole stack kept to two ~300-line files: train.py for the loop and model.py for the architecture. A char-level Shakespeare run finishes in ~3 minutes on one GPU.
AI Train2023
Modular PyTorch-based framework for building, training, and deploying physics-informed ML models (neural operators, PINNs, GNNs, diffusion). Provides GPU‑optimized training, domain-specific datapipes for meshes/point clouds, distributed scaling and a model zoo.
AI Train2023
Open platform for training, serving, and evaluating LLM chatbots; ships a distributed multi-model serving system with OpenAI-compatible APIs. Release home of Vicuna and Chatbot Arena, whose 1.5M+ human votes power an Elo leaderboard across 70+ models.
AI Train2023
Streamlines post-training and fine-tuning for large language and multimodal models with a single YAML-driven pipeline. Supports LoRA/QLoRA, full fine-tuning, preference tuning, RL methods, multi-GPU/FSDP/DeepSpeed, and many model backends (Hugging Face, local checkpoints).
AI Train2023
Connects a frozen vision encoder to a language model via visual instruction tuning, yielding an open multimodal assistant that follows image-grounded instructions. Released checkpoints span 7B-34B and approach GPT-4V on vision-language benchmarks.