Qwen Code is a coding agent that lives in the digital world.

Dion optimizer algorithm

Democratizing Reinforcement Learning for LLMs

This repository implements a robust Min-Max Nonlinear Model Predictive Control (NMPC) approach for tracking control of nonlinear systems subject to bounded disturbances

Lean 4 programming language and theorem prover

A unified library of state-of-the-art model optimization techniques like quantization, pruning, distillation, speculative decoding, etc. It compresses deep learning models for downstream deployment…

Solve and estimate Dynamic Stochastic General Equilibrium models (including the New York Fed DSGE)

Fine-tuning & Reinforcement Learning for LLMs. 🦥 Train OpenAI gpt-oss, DeepSeek-R1, Qwen3, Gemma 3, TTS 2x faster with 70% less VRAM.

🍒 Cherry Studio is a desktop client that supports for multiple LLM providers.

Learn Lean 4 with PLFA proofs.

OpenAI Frontier Evals

Comparison between GFlowNets & Maximum Entropy RL

Question paper of courses taught at IISC as part of MTech AI curriculum

Official repository of "SAMURAI: Adapting Segment Anything Model for Zero-Shot Visual Tracking with Motion-Aware Memory"

Official repository for the Boltz biomolecular interaction models

AI agent with RAG+ReAct on Indian Constitution & BNS

An introduction to LLM Sampling

For optimization algorithm research and development.

Sidecar is the AI brains for the Aide editor and works alongside it, locally on your machine

Large Reasoning Models

OCR & Document Extraction using vision models

Entropy Based Sampling and Parallel CoT Decoding

Large-scale Self-supervised Pre-training Across Tasks, Languages, and Modalities