haoliuhl / ringattention
Transformers with Arbitrarily Large Context
☆637Updated 3 months ago
Related projects ⓘ
Alternatives and complementary repositories for ringattention
- Implementation of 💍 Ring Attention, from Liu et al. at Berkeley AI, in Pytorch☆474Updated 2 weeks ago
- Ring attention implementation with flash attention☆578Updated this week
- [ICML 2024] Break the Sequential Dependency of LLM Inference Using Lookahead Decoding☆1,141Updated 3 weeks ago
- Microsoft Automatic Mixed Precision Library☆522Updated last month
- [ICLR 2024] Sheared LLaMA: Accelerating Language Model Pre-training via Structured Pruning☆553Updated 8 months ago
- Official PyTorch implementation of QA-LoRA☆116Updated 7 months ago
- Memory optimization and training recipes to extrapolate language models' context length to 1 million tokens, with minimal hardware.☆644Updated last month
- [NeurIPS 2024] KVQuant: Towards 10 Million Context Length LLM Inference with KV Cache Quantization☆303Updated 2 months ago
- Helpful tools and examples for working with flex-attention☆462Updated 2 weeks ago
- Scalable toolkit for efficient model alignment☆614Updated this week
- [ICML 2024] CLLMs: Consistency Large Language Models☆351Updated this week
- Code for paper: "QuIP: 2-Bit Quantization of Large Language Models With Guarantees"☆347Updated 8 months ago
- Explorations into some recent techniques surrounding speculative decoding☆210Updated last year
- USP: Unified (a.k.a. Hybrid, 2D) Sequence Parallel Attention for Long Context Transformers Model Training and Inference☆353Updated last week
- 🚀 Efficiently (pre)training foundation models with native PyTorch features, including FSDP for training and SDPA implementation of Flash…☆191Updated 3 weeks ago
- Implementation of ST-Moe, the latest incarnation of MoE after years of research at Brain, in Pytorch☆291Updated 4 months ago
- Official repository for LightSeq: Sequence Level Parallelism for Distributed Training of Long Context Transformers☆196Updated 2 months ago
- Pipeline Parallelism for PyTorch☆725Updated 2 months ago
- [ICML 2024] SqueezeLLM: Dense-and-Sparse Quantization☆643Updated 2 months ago
- FP16xINT4 LLM inference kernel that can achieve near-ideal ~4x speedups up to medium batchsizes of 16-32 tokens.☆611Updated 2 months ago
- ☆284Updated 7 months ago
- A simple and effective LLM pruning approach.☆660Updated 3 months ago
- Triton-based implementation of Sparse Mixture of Experts.☆184Updated last month
- Tutel MoE: An Optimized Mixture-of-Experts Implementation☆729Updated 2 weeks ago
- Official Implementation of EAGLE-1 (ICML'24) and EAGLE-2 (EMNLP'24)☆817Updated last week
- distributed trainer for LLMs☆543Updated 5 months ago
- This repository contains the experimental PyTorch native float8 training UX☆211Updated 3 months ago
- PyTorch implementation of Infini-Transformer from "Leave No Context Behind: Efficient Infinite Context Transformers with Infini-attention…☆279Updated 6 months ago
- Minimalistic large language model 3D-parallelism training☆1,229Updated last week