linfeng93 / BiTA
An innovative method expediting LLMs via streamlined semi-autoregressive generation and draft verification.
☆25Updated this week
Alternatives and similar repositories for BiTA:
Users that are interested in BiTA are comparing it to the libraries listed below
- [NeurIPS 2024] The official implementation of "Kangaroo: Lossless Self-Speculative Decoding for Accelerating LLMs via Double Early Exitin…☆51Updated 9 months ago
- Ouroboros: Speculative Decoding with Large Model Enhanced Drafting (EMNLP 2024 main)☆101Updated last month
- ☆122Updated 2 months ago
- ☆68Updated 4 months ago
- QAQ: Quality Adaptive Quantization for LLM KV Cache☆49Updated last year
- GEAR: An Efficient KV Cache Compression Recipefor Near-Lossless Generative Inference of LLM☆159Updated 9 months ago
- Boosting 4-bit inference kernels with 2:4 Sparsity☆72Updated 7 months ago
- Multi-Candidate Speculative Decoding☆35Updated 11 months ago
- [MLSys'24] Atom: Low-bit Quantization for Efficient and Accurate LLM Serving☆303Updated 9 months ago
- ☆43Updated last year
- Implementation of Speculative Sampling as described in "Accelerating Large Language Model Decoding with Speculative Sampling" by Deepmind☆93Updated last year
- Odysseus: Playground of LLM Sequence Parallelism☆68Updated 10 months ago
- Sirius, an efficient correction mechanism, which significantly boosts Contextual Sparsity models on reasoning tasks while maintaining its…☆21Updated 7 months ago
- [ICLR 2025] PEARL: Parallel Speculative Decoding with Adaptive Draft Length☆75Updated this week
- [ICLR2025] Breaking Throughput-Latency Trade-off for Long Sequences with Speculative Decoding☆113Updated 4 months ago
- Repository for Sparse Finetuning of LLMs via modified version of the MosaicML llmfoundry☆40Updated last year
- ☆237Updated 11 months ago
- PyTorch bindings for CUTLASS grouped GEMM.☆81Updated 5 months ago
- 16-fold memory access reduction with nearly no loss☆89Updated 3 weeks ago
- ☆20Updated this week
- Code associated with the paper **Draft & Verify: Lossless Large Language Model Acceleration via Self-Speculative Decoding**☆181Updated 2 months ago
- Code for paper: [ICLR2025 Oral] FlexPrefill: A Context-Aware Sparse Attention Mechanism for Efficient Long-Sequence Inference☆85Updated this week
- Official PyTorch implementation of "IntactKV: Improving Large Language Model Quantization by Keeping Pivot Tokens Intact"☆43Updated 10 months ago
- An easy-to-use package for implementing SmoothQuant for LLMs☆96Updated last week
- [NeurIPS'23] Speculative Decoding with Big Little Decoder☆90Updated last year
- ☆68Updated 2 months ago
- Vocabulary Parallelism☆17Updated last month
- ☆48Updated 4 months ago
- [ICLR 2025] TidalDecode: A Fast and Accurate LLM Decoding with Position Persistent Sparse Attention☆33Updated 2 weeks ago
- Official PyTorch implementation of "FlatQuant: Flatness Matters for LLM Quantization"☆117Updated 3 weeks ago