Stars
Fully open reproduction of DeepSeek-R1
Survey and paper list on efficiency-guided LLM agents (memory, tool learning, planning).
😎 A Survey of Efficient Reasoning for Large Reasoning Models: Language, Multimodality, Agent, and Beyond
kill that logs papers to Google Sheets — just say "add this paper" or paste an arXiv URL.
Minimalistic 4D-parallelism distributed training framework for education purpose
Turn complex codebases into clear, navigable architecture diagrams with Claude Code.
Teams-first Multi-agent orchestration for Claude Code
DeepGEMM: clean and efficient FP8 GEMM kernels with fine-grained scaling
Implementation for FP8/INT8 Rollout for RL training without performence drop.
NVIDIA Resiliency Extension is a python package for framework developers and users to implement fault-tolerant features. It improves the effective training time by minimizing the downtime due to fa…
[ICLR 2025] Official PyTorch Implementation of Gated Delta Networks: Improving Mamba2 with Delta Rule
어르신계정 password: 1234 (가족 계정: family1/password123)
Ongoing research training transformer models at scale
43 tips for getting the most out of Claude Code, from basics to advanced - includes a custom status line script and Claude Code running itself in a container. Also includes the dx plugin.
MinHash, LSH, LSH Forest, Weighted MinHash, HyperLogLog, HyperLogLog++, LSH Ensemble and HNSW
Freeing data processing from scripting madness by providing a set of platform-agnostic customizable pipeline processing blocks.
Data processing for and with foundation models! 🍎 🍋 🌽 ➡️ ➡️🍸 🍹 🍷
Ray is an AI compute engine. Ray consists of a core distributed runtime and a set of AI Libraries for accelerating ML workloads.
Benchmarking Complex Instruction-Following with Multiple Constraints Composition (NeurIPS 2024 Datasets and Benchmarks Track)
Minimal Claude Code alternative. Single Python file, zero dependencies, ~250 lines.
<밑바닥부터 시작하는 딥러닝 4>