NVIDIA / Star-AttentionLinks
Efficient LLM Inference over Long Sequences
β378Updated 2 weeks ago
Alternatives and similar repositories for Star-Attention
Users that are interested in Star-Attention are comparing it to the libraries listed below
Sorting:
- LLM KV cache compression made easyβ520Updated this week
- π Efficiently (pre)training foundation models with native PyTorch features, including FSDP for training and SDPA implementation of Flashβ¦β253Updated this week
- A throughput-oriented high-performance serving framework for LLMsβ825Updated 2 weeks ago
- [ICLR 2025] DuoAttention: Efficient Long-Context LLM Inference with Retrieval and Streaming Headsβ466Updated 4 months ago
- [MLSys'25] QServe: W4A8KV4 Quantization and System Co-design for Efficient LLM Serving; [MLSys'25] LServe: Efficient Long-sequence LLM Seβ¦β705Updated 3 months ago
- Towards Economical Inference: Enabling DeepSeek's Multi-Head Latent Attention in Any Transformer-based LLMsβ176Updated this week
- [NeurIPS 2024] KVQuant: Towards 10 Million Context Length LLM Inference with KV Cache Quantizationβ359Updated 10 months ago
- Code for Neurips24 paper: QuaRot, an end-to-end 4-bit inference of large language models.β395Updated 6 months ago
- [MLSys'24] Atom: Low-bit Quantization for Efficient and Accurate LLM Servingβ310Updated 11 months ago
- Advanced Quantization Algorithm for LLMs and VLMs, with support for CPU, Intel GPU, CUDA and HPU. Seamlessly integrated with Torchao, Traβ¦β525Updated this week
- [ACL 2025 Main] EfficientQAT: Efficient Quantization-Aware Training for Large Language Modelsβ273Updated last month
- [ICML 2024] KIVI: A Tuning-Free Asymmetric 2bit Quantization for KV Cacheβ303Updated 5 months ago
- KV cache compression for high-throughput LLM inferenceβ131Updated 4 months ago
- Applied AI experiments and examples for PyTorchβ277Updated 3 weeks ago
- [ICLR'25] Fast Inference of MoE Models with CPU-GPU Orchestrationβ213Updated 7 months ago
- β159Updated this week
- β256Updated last year
- [ICLR 2025] COAT: Compressing Optimizer States and Activation for Memory-Efficient FP8 Trainingβ209Updated last week
- Code for "LayerSkip: Enabling Early Exit Inference and Self-Speculative Decoding", ACL 2024β311Updated last month
- [NeurIPS'24 Spotlight, ICLR'25, ICML'25] To speed up Long-context LLMs' inference, approximate and dynamic sparse calculate the attentionβ¦β1,055Updated last week
- Fast low-bit matmul kernels in Tritonβ322Updated this week
- Repo for "LoLCATs: On Low-Rank Linearizing of Large Language Models"β239Updated 4 months ago
- USP: Unified (a.k.a. Hybrid, 2D) Sequence Parallel Attention for Long Context Transformers Model Training and Inferenceβ519Updated 3 weeks ago
- Ring attention implementation with flash attentionβ789Updated last week
- Perplexity GPU Kernelsβ375Updated 2 weeks ago
- π³ Efficient Triton implementations for "Native Sparse Attention: Hardware-Aligned and Natively Trainable Sparse Attention"β702Updated 3 months ago
- β193Updated 4 months ago
- Efficient triton implementation of Native Sparse Attention.β168Updated last month
- An efficient GPU support for LLM inference with x-bit quantization (e.g. FP6,FP5).β252Updated 7 months ago
- kernels, of the mega varietyβ390Updated 3 weeks ago