Infini-AI-Lab / Sequoia
scalable and robust tree-based speculative decoding algorithm
☆336Updated last month
Alternatives and similar repositories for Sequoia:
Users that are interested in Sequoia are comparing it to the libraries listed below
- ☆528Updated 4 months ago
- [ICML 2024] CLLMs: Consistency Large Language Models☆385Updated 3 months ago
- Code for the paper "QMoE: Practical Sub-1-Bit Compression of Trillion-Parameter Models".☆270Updated last year
- REST: Retrieval-Based Speculative Decoding, NAACL 2024☆197Updated 3 months ago
- Explorations into some recent techniques surrounding speculative decoding☆246Updated 2 months ago
- [MLSys'24] Atom: Low-bit Quantization for Efficient and Accurate LLM Serving☆298Updated 8 months ago
- Experiments on speculative sampling with Llama models☆125Updated last year
- [NeurIPS 2024] KVQuant: Towards 10 Million Context Length LLM Inference with KV Cache Quantization☆335Updated 7 months ago
- Code for paper: "QuIP: 2-Bit Quantization of Large Language Models With Guarantees"☆362Updated last year
- [ICLR'25] Fast Inference of MoE Models with CPU-GPU Orchestration☆198Updated 3 months ago
- FP16xINT4 LLM inference kernel that can achieve near-ideal ~4x speedups up to medium batchsizes of 16-32 tokens.☆755Updated 6 months ago
- ☆193Updated 3 months ago
- [ICLR2025] Breaking Throughput-Latency Trade-off for Long Sequences with Speculative Decoding☆109Updated 3 months ago
- Fast Matrix Multiplications for Lookup Table-Quantized LLMs☆231Updated 2 weeks ago
- ☆511Updated 6 months ago
- [ICLR 2025] DuoAttention: Efficient Long-Context LLM Inference with Retrieval and Streaming Heads☆431Updated last month
- [MLSys'25] QServe: W4A8KV4 Quantization and System Co-design for Efficient LLM Serving; [MLSys'25] LServe: Efficient Long-sequence LLM Se…☆594Updated last week
- ☆229Updated 10 months ago
- Layer-Condensed KV cache w/ 10 times larger batch size, fewer params and less computation. Dramatic speed up with better task performance…☆147Updated last month
- Official repository for LightSeq: Sequence Level Parallelism for Distributed Training of Long Context Transformers☆207Updated 6 months ago
- An efficent implementation of the method proposed in "The Era of 1-bit LLMs"☆154Updated 4 months ago
- Repository for the QUIK project, enabling the use of 4bit kernels for generative inference - EMNLP 2024☆177Updated 10 months ago
- ☆310Updated 11 months ago
- Implementation of Speculative Sampling as described in "Accelerating Large Language Model Decoding with Speculative Sampling" by Deepmind☆90Updated last year
- EfficientQAT: Efficient Quantization-Aware Training for Large Language Models☆250Updated 5 months ago
- [ICML 2024] KIVI: A Tuning-Free Asymmetric 2bit Quantization for KV Cache☆279Updated last month
- ☆115Updated 3 weeks ago
- Unofficial implementation for the paper "Mixture-of-Depths: Dynamically allocating compute in transformer-based language models"☆152Updated 8 months ago
- KV cache compression for high-throughput LLM inference☆118Updated last month
- LLM KV cache compression made easy☆428Updated last week