dame-cell / Triformer
Transformers components but in Triton
☆25Updated this week
Related projects ⓘ
Alternatives and complementary repositories for Triformer
- ☆22Updated 10 months ago
- Odysseus: Playground of LLM Sequence Parallelism☆55Updated 4 months ago
- Quantized Attention on GPU☆29Updated last week
- A Suite for Parallel Inference of Diffusion Transformers (DiTs) on multi-GPU Clusters☆32Updated 3 months ago
- Boosting 4-bit inference kernels with 2:4 Sparsity☆51Updated 2 months ago
- PyTorch bindings for CUTLASS grouped GEMM.☆53Updated last week
- The official implementation of paper: SimLayerKV: A Simple Framework for Layer-Level KV Cache Reduction.☆36Updated 3 weeks ago
- Code for Palu: Compressing KV-Cache with Low-Rank Projection☆56Updated this week
- Ouroboros: Speculative Decoding with Large Model Enhanced Drafting (EMNLP 2024 main)☆76Updated 3 weeks ago
- NAACL '24 (Best Demo Paper RunnerUp) / MlSys @ NeurIPS '23 - RedCoast: A Lightweight Tool to Automate Distributed Training and Inference☆60Updated 3 weeks ago
- GPTQ inference TVM kernel☆35Updated 6 months ago
- MagicPIG: LSH Sampling for Efficient LLM Generation☆45Updated 2 weeks ago
- [ACL 2024] RelayAttention for Efficient Large Language Model Serving with Long System Prompts☆34Updated 8 months ago
- TensorRT LLM Benchmark Configuration☆11Updated 3 months ago
- Sequence-level 1F1B schedule for LLMs.☆17Updated 5 months ago
- 16-fold memory access reduction with nearly no loss☆57Updated this week
- Pytorch implementation of our paper accepted by ICML 2024 -- CaM: Cache Merging for Memory-efficient LLMs Inference☆25Updated 4 months ago
- ☆42Updated 6 months ago
- ☆55Updated 5 months ago
- GEAR: An Efficient KV Cache Compression Recipefor Near-Lossless Generative Inference of LLM☆146Updated 4 months ago
- ☆29Updated 5 months ago
- Stick-breaking attention☆33Updated this week
- Materials for learning SGLang☆86Updated this week
- CUDA and Triton implementations of Flash Attention with SoftmaxN.☆66Updated 5 months ago
- Breaking Throughput-Latency Trade-off for Long Sequences with Speculative Decoding☆70Updated last week
- Fast and Robust Early-Exiting Framework for Autoregressive Language Models with Synchronized Parallel Decoding (EMNLP 2023 Long)☆52Updated last month
- Implementation of Kangaroo: Lossless Self-Speculative Decoding via Double Early Exiting☆41Updated 4 months ago
- Puzzles for learning Triton, play it with minimal environment configuration!☆61Updated this week
- ☆95Updated last month