void-main / FasterTransformer
Transformer related optimization, including BERT, GPT
☆60Updated last year
Related projects ⓘ
Alternatives and complementary repositories for FasterTransformer
- ☆140Updated 6 months ago
- ☆123Updated 2 weeks ago
- Performance of the C++ interface of flash attention and flash attention v2 in large language model (LLM) inference scenarios.☆29Updated 2 months ago
- An easy-to-use package for implementing SmoothQuant for LLMs☆83Updated 6 months ago
- A collection of memory efficient attention operators implemented in the Triton language.☆219Updated 5 months ago
- PyTorch bindings for CUTLASS grouped GEMM.☆68Updated 4 months ago
- ☆138Updated 2 weeks ago
- QQQ is an innovative and hardware-optimized W4A8 quantization solution for LLMs.☆87Updated last month
- ☆74Updated 11 months ago
- ☆79Updated 2 months ago
- [USENIX ATC '24] Accelerating the Training of Large Language Models using Efficient Activation Rematerialization and Optimal Hybrid Paral…☆45Updated 3 months ago
- Transformer related optimization, including BERT, GPT☆39Updated last year
- ☆21Updated last year
- ☆64Updated 3 months ago
- export llama to onnx☆96Updated 5 months ago
- Running BERT without Padding☆460Updated 2 years ago
- ☆99Updated 8 months ago
- Dynamic Memory Management for Serving LLMs without PagedAttention☆238Updated last week
- Train llm (bloom, llama, baichuan2-7b, chatglm3-6b) with deepspeed pipeline mode. Faster than zero/zero++/fsdp.☆90Updated 9 months ago
- [ACL 2024] A novel QAT with Self-Distillation framework to enhance ultra low-bit LLMs.☆84Updated 6 months ago
- A standalone GEMM kernel for fp16 activation and quantized weight, extracted from FasterTransformer☆85Updated 8 months ago
- Summary of system papers/frameworks/codes/tools on training or serving large model☆56Updated 11 months ago
- ☆57Updated 2 weeks ago
- Compare different hardware platforms via the Roofline Model for LLM inference tasks.☆75Updated 8 months ago
- Simple Dynamic Batching Inference☆145Updated 2 years ago
- USP: Unified (a.k.a. Hybrid, 2D) Sequence Parallel Attention for Long Context Transformers Model Training and Inference☆357Updated this week
- flash attention tutorial written in python, triton, cuda, cutlass☆202Updated 5 months ago
- optimized BERT transformer inference on NVIDIA GPU. https://arxiv.org/abs/2210.03052☆457Updated 8 months ago
- ☆111Updated 8 months ago
- This repository contains integer operators on GPUs for PyTorch.☆183Updated last year