#Flash-attention

Showing 13 of 13 repositories tagged #flash-attention, ranked by stars

QwenLM
QwenLM
Qwen

The official repo of Qwen (通义千问) chat & pretrained large language model proposed by Alibaba Cloud.

Score
100
★ 21.4k ⑂ 1.8k +17/day
Python
InternLM
InternLM
InternLM

Official release of InternLM series (InternLM, InternLM2, InternLM2.5, InternLM3).

Score
25
★ 7.2k ⑂ 508
Python
ymcui
ymcui
Chinese-LLaMA-Alpaca-2

中文LLaMA-2 & Alpaca-2大模型二期项目 + 64K超长上下文模型 (Chinese LLaMA-2 & Alpaca-2 LLMs with 64K long context models)

Score
75
★ 7.1k ⑂ 564
Python
xlite-dev
xlite-dev
Awesome-LLM-Inference

📚A curated list of Awesome LLM/VLM Inference Papers with Codes: Flash-Attention, Paged-Attention, WINT8/4, Parallelism, etc.🎉

Score
50
★ 5.4k ⑂ 420 +19/day
Python
MoonshotAI
MoonshotAI
MoBA

MoBA: Mixture of Block Attention for Long-Context LLMs

Score
100
★ 2.1k ⑂ 154 +2/day
Python
NVIDIA
NVIDIA
cudnn-frontend

cuDNN Frontend is NVIDIA's modern, open-source entry point to the cuDNN library and a growing collection of high-performance open-source kernels.

Score
0
★ 864 ⑂ 200 +2/day
Python
HKUSTDial
HKUSTDial
flash-sparse-attention

Trainable fast and memory-efficient sparse attention

Score
0
★ 728 ⑂ 52 +9/day
Python
zengxiao-he
zengxiao-he
tessera

From teacher to tiles — a from-scratch LLM distillation & serving engine: custom Triton/CUDA kernels, FSDP distillation, paged-KV continuous batching, speculative decoding, a Rust gateway, a JAX oracle, and interpretability tooling.

Score
0
★ 512 ⑂ 5 +24/day
Python
windreamer
windreamer
flash-attention3-wheels

Pre-built wheels that erase Flash Attention 3 installation headaches.

Score
80
★ 101 ⑂ 6
Python
CoinCheung
CoinCheung
gdGPT

Train llm (bloom, llama, baichuan2-7b, chatglm3-6b) with deepspeed pipeline mode. Faster than zero/zero++/fsdp.

Score
40
★ 97 ⑂ 10
Python
DaoyuanLi2816
DaoyuanLi2816
llama3-emotion-lora

Emotion text classification using Llama3-8b with LoRA and FlashAttention. Based on LLaMA-Factory.

Score
60
★ 73 ⑂ 10
Python
Bruce-Lee-LY
Bruce-Lee-LY
decoding_attention

Decoding Attention is specially optimized for MHA, MQA, GQA and MLA using CUDA core for the decoding stage of LLM inference.

Score
0
★ 47 ⑂ 4
C++
Bruce-Lee-LY
Bruce-Lee-LY
flash_attention_inference

Performance of the C++ interface of flash attention and flash attention v2 in large language model (LLM) inference scenarios.

Score
20
★ 45 ⑂ 7
C++
Related Topics
#llm#cuda#large-language-model#nlp#pytorch#gpu#nvidia#chinese#large-language-models#pretrained-models#rlhf#llama2

© 2026 GitRepoTrend · GitHub repositories by topic · Updated weekly