[CVPR 2025] VGGT: Visual Geometry Grounded Transformer

EvolKit is an innovative framework designed to automatically enhance the complexity of instructions used for fine-tuning Large Language Models (LLMs).

ArcticTraining is a framework designed to simplify and accelerate the post-training process for large language models (LLMs)

Website for CS 265

A community-maintained Python framework for creating mathematical animations.

Animation engine for explanatory math videos

XLB: Accelerated Lattice Boltzmann (XLB) for Physics-based ML

🐈 Automatic differentiable spectrum modeling of exoplanets/brown dwarfs using JAX, compatible with NumPyro and Optax/JAXopt

Bayesian Neural Field models for prediction in large-scale spatiotemporal datasets

Robust recipes to align language models with human and AI preferences

Algorithmically create or extend categorical colour palettes

Train high-quality text-to-image diffusion models in a data & compute efficient manner

A series of math-specific large language models of our Qwen2 series.

Sparsify transformers with SAEs and transcoders

Distribute and run LLMs with a single file.

Creates performance portable libraries with embedded source representations.

Simplifying reinforcement learning for complex game environments

MiniCPM-o 2.6: A GPT-4o Level MLLM for Vision, Speech and Multimodal Live Streaming on Your Phone

arXiv LaTeX Cleaner: Easily clean the LaTeX code of your paper to submit to arXiv

Gemma 2B with 10M context length using Infini-attention.

Utilities for constructing a large dataset of LLVM IR

A PyTorch native library for large model training

Grammars written for ANTLR v4; expectation that the grammars are free of actions.

Open weights language model from Google DeepMind, based on Griffin.

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

Go ahead and axolotl questions

Scientific computing in Lean 4

🌲 A 3D, interactive semantic graph of hacker interests at TreeHacks, scraped from Slack intro messages