Lists (32)
AI Efficiency
482 repositories
Attention
56 repositories
Checkpointing
29 repositories
Collective Communication
52 repositories
Compilers
132 repositories
CUDA
236 repositories
Datasets
55 repositories
Diffusion
241 repositories
Distributed Systems
170 repositories
Graphs
13 repositories
gRPC
20 repositories
Hardware
81 repositories
HPC
92 repositories
Infrastructure as Code
30 repositories
Jax
50 repositories
Job Orchestration
20 repositories
K8s
44 repositories
Kernel Language
145 repositories
Language Models
502 repositories
Linux
34 repositories
Mechanistic Interpretability
31 repositories
NVIDIA GDS: DMA/RDMA
36 repositories
PEFT
16 repositories
PyTorch
279 repositories
Quantization
55 repositories
Research Tools
283 repositories
Rust Machine Learning Ecosystem
40 repositories
Serving
178 repositories
Simulation
47 repositories
Storage
20 repositories
Vision
88 repositories
WASM
27 repositories
Starred repositories
MoE training system for research, speed-running and profit
Parrot is a C++ library for fused array operations using CUDA/Thrust. It provides efficient GPU-accelerated operations with lazy evaluation semantics, allowing for chaining of operations without un…
Cosmos-Predict2.5, the latest version of the Cosmos World Foundation Models (WFMs) family, specialized for simulating and predicting the future state of the world in the form of video.
easily compare pytorch json profiles. vibe coded
Zonos-v0.1 is a leading open-weight text-to-speech model trained on more than 200k hours of varied multilingual speech, delivering expressiveness and quality on par with—or even surpassing—top TTS …
Technical report of Kimina-Prover Preview.
DC-Gen: Post-Training Diffusion Acceleration with Deeply Compressed Latent Space
Scalable, fast, and disk-friendly vector search in Postgres, the successor of pgvecto.rs.
AIPerf is a comprehensive benchmarking tool that measures the performance of generative AI models served by your preferred inference solution.
An algorithm for weight-activation quantization (W4A4, W4A8) of LLMs, supporting both static and dynamic quantization
Post-training with Tinker
Github mirror of trition-lang/triton repo.
Kimi K2 is the large language model series developed by Moonshot AI team
Home for "How To Scale Your Model", a short blog-style textbook about scaling LLMs on TPUs
Verify Precision of all Kimi K2 API Vendor
A minimal implementation of DeepMind's Genie world model
Research code artifacts for Code World Model (CWM) including inference tools, reproducibility, and documentation.