NVIDIA / Megatron-Energon
Megatron's multi-modal data loader
☆157Updated this week
Alternatives and similar repositories for Megatron-Energon:
Users that are interested in Megatron-Energon are comparing it to the libraries listed below
- This repository contains the experimental PyTorch native float8 training UX☆219Updated 5 months ago
- 🚀 Efficiently (pre)training foundation models with native PyTorch features, including FSDP for training and SDPA implementation of Flash…☆215Updated this week
- LLM KV cache compression made easy☆303Updated this week
- Official repository for LightSeq: Sequence Level Parallelism for Distributed Training of Long Context Transformers☆204Updated 4 months ago
- Applied AI experiments and examples for PyTorch☆211Updated this week
- Breaking Throughput-Latency Trade-off for Long Sequences with Speculative Decoding☆107Updated last month
- Efficient LLM Inference over Long Sequences☆344Updated 3 weeks ago
- Fast Matrix Multiplications for Lookup Table-Quantized LLMs☆219Updated this week
- 🚀 Collection of components for development, training, tuning, and inference of foundation models leveraging PyTorch native components.☆182Updated this week
- KV cache compression for high-throughput LLM inference☆103Updated last month
- Triton-based implementation of Sparse Mixture of Experts.☆192Updated last month
- Model Compression Toolbox for Large Language Models and Diffusion Models☆302Updated 3 weeks ago
- ring-attention experiments☆116Updated 3 months ago
- Tritonbench is a collection of PyTorch custom operators with example inputs to measure their performance.☆75Updated this week
- [ICML 2024] Quest: Query-Aware Sparsity for Efficient Long-Context LLM Inference☆234Updated last month
- USP: Unified (a.k.a. Hybrid, 2D) Sequence Parallel Attention for Long Context Transformers Model Training and Inference☆403Updated 2 weeks ago
- Boosting 4-bit inference kernels with 2:4 Sparsity☆64Updated 4 months ago
- Zero Bubble Pipeline Parallelism☆309Updated 2 months ago
- PyTorch bindings for CUTLASS grouped GEMM.☆84Updated 2 weeks ago
- DuoAttention: Efficient Long-Context LLM Inference with Retrieval and Streaming Heads☆417Updated 2 months ago
- ☆212Updated 8 months ago
- Ring attention implementation with flash attention☆647Updated last month
- ☆109Updated last week
- ☆170Updated last week
- The official implementation of the paper <MoA: Mixture of Sparse Attention for Automatic Large Language Model Compression>☆110Updated last month
- GEAR: An Efficient KV Cache Compression Recipefor Near-Lossless Generative Inference of LLM☆152Updated 6 months ago
- Implementation of 💍 Ring Attention, from Liu et al. at Berkeley AI, in Pytorch☆492Updated 2 months ago
- Fast low-bit matmul kernels in Triton☆187Updated last week
- [NeurIPS 2024] KVQuant: Towards 10 Million Context Length LLM Inference with KV Cache Quantization☆325Updated 5 months ago
- [MLSys'24] Atom: Low-bit Quantization for Efficient and Accurate LLM Serving☆290Updated 6 months ago