Accelerating Large-Scale Reasoning Model Inference with Sparse Self-Speculative Decoding
☆122Dec 2, 2025Updated 10 months ago
Alternatives and similar repositories for SparseSpec
Users that are interested in SparseSpec are comparing it to the libraries listed below. We may earn a commission when you buy through links labeled 'Ad' on this page.
Sorting:
- Vortex: Programmable Sparse Attention for Agents as Algorithm Designers☆69Aug 22, 2026Updated 2 months ago
- [NeurIPS'25 Spotlight] Adaptive Attention Sparsity with Hierarchical Top-p Pruning☆110Aug 31, 2026Updated last month
- [ASPLOS'26] Taming the Long-Tail: Efficient Reasoning RL Training with Adaptive Drafter☆186Feb 27, 2026Updated 8 months ago
- [ICLR2025] Breaking Throughput-Latency Trade-off for Long Sequences with Speculative Decoding☆161Dec 4, 2024Updated last year
- [ICLR 2025] TidalDecode: A Fast and Accurate LLM Decoding with Position Persistent Sparse Attention☆57Aug 6, 2025Updated last year
- Serverless GPU API endpoints on Runpod - Get Bonus Credits • AdSkip the infrastructure headaches. Auto-scaling, pay-as-you-go, no-ops approach lets you focus on innovating your application.
- ☆255Nov 19, 2025Updated 11 months ago
- Code for the papers: “Four Over Six: More Accurate NVFP4 Quantization with Adaptive Block Scaling” and “Adaptive Block-Scaled Data Types”☆209Apr 21, 2026Updated 6 months ago
- SC'25 UltraAttn: Efficiently Parallelizing Attention through Hierarchical Context-Tiling☆16Aug 14, 2025Updated last year
- [COLM 2024] TriForce: Lossless Acceleration of Long Sequence Generation with Hierarchical Speculative Decoding☆280Aug 31, 2024Updated 2 years ago
- [MLSys'24] Atom: Low-bit Quantization for Efficient and Accurate LLM Serving☆350Jul 2, 2024Updated 2 years ago
- Code repository for the SOSP'25 paper DCP: Addressing Input Dynamism In Long-Context Training via Dynamic Context Parallelism.☆23Nov 28, 2025Updated 11 months ago
- Fast, memory-efficient attention column reduction (e.g., sum, mean, max)☆50Feb 10, 2026Updated 8 months ago
- Accelerating MoE with IO and Tile-aware Optimizations☆784Aug 29, 2026Updated 2 months ago
- [ICML 2026] Less Is More: Training-Free Sparse Attention with Global Locality for Efficient Reasoning☆36Sep 12, 2025Updated last year
- Deploy on Railway without the complexity - Free Credits Offer • AdConnect your repo and Railway handles the rest with instant previews. Quickly provision container image services, databases, and storage volumes.
- ArkVale: Efficient Generative LLM Inference with Recallable Key-Value Eviction (NIPS'24)☆54Dec 17, 2024Updated last year
- Prefix-Aware Attention for LLM Decoding☆48May 26, 2026Updated 5 months ago
- [ICLR 2025] DuoAttention: Efficient Long-Context LLM Inference with Retrieval and Streaming Heads☆544Feb 10, 2025Updated last year
- A WebUI for Side-by-Side Comparison of Media (Images/Videos) Across Multiple Folders☆26Feb 21, 2025Updated last year
- [EMNLP2026 Main] Official implementation of “Domino: Decoupling Causal Modeling from Autoregressive Drafting in Speculative Decoding”.☆139Jul 25, 2026Updated 3 months ago
- PoC for "SpecReason: Fast and Accurate Inference-Time Compute via Speculative Reasoning" [NeurIPS '25]☆76Oct 2, 2025Updated last year
- [ICML 2024] Quest: Query-Aware Sparsity for Efficient Long-Context LLM Inference☆414Jul 10, 2025Updated last year
- Draft-Target Disaggregation LLM Serving System via Parallel Speculative Decoding.☆224Mar 18, 2026Updated 7 months ago
- A Survey of Efficient Attention Methods: Hardware-efficient, Sparse, Compact, and Linear Attention☆313Dec 1, 2025Updated 10 months ago
- Deploy on Railway without the complexity - Free Credits Offer • AdConnect your repo and Railway handles the rest with instant previews. Quickly provision container image services, databases, and storage volumes.
- [NeurIPS 2025] Speculate Deep and Accurate☆25Aug 10, 2026Updated 2 months ago
- Preview Code for Continuum Paper☆112Aug 13, 2026Updated 2 months ago
- [MLSys'25] QServe: W4A8KV4 Quantization and System Co-design for Efficient LLM Serving; [MLSys'25] LServe: Efficient Long-sequence LLM Se…☆867Mar 6, 2025Updated last year
- mKernel: fast multi-node, multi-GPU fused kernels☆305Oct 17, 2026Updated last week
- ☆24Mar 3, 2026Updated 7 months ago
- 16-fold memory access reduction with nearly no loss☆107Mar 26, 2025Updated last year
- [ICLR2025 Spotlight] MagicPIG: LSH Sampling for Efficient LLM Generation☆257Dec 16, 2024Updated last year
- [ICLR 2026] Draft-based Approximate Inference for LLMs☆24Mar 10, 2026Updated 7 months ago
- Research prototype of PRISM — a cost-efficient multi-LLM serving system with flexible time- and space-based GPU sharing.☆77Mar 17, 2026Updated 7 months ago
- Proton VPN Special Offer - Get 70% off • AdSpecial partner offer. Trusted by over 100 million users worldwide. Tested, Approved and Recommended by Experts.
- [ICLR 2025] Palu: Compressing KV-Cache with Low-Rank Projection☆166Feb 20, 2025Updated last year
- [ICLR 2025] DeFT: Decoding with Flash Tree-attention for Efficient Tree-structured LLM Inference☆57Jun 17, 2025Updated last year
- Accepted to MLSys 2026☆97Apr 19, 2026Updated 6 months ago
- ☆82Aug 21, 2024Updated 2 years ago
- Kernel Design Agents (KDA) is a agent-centric workflow to write high-performance CUDA Kernels.☆1,464Oct 19, 2026Updated last week
- UniQL official repository (ICLR 2026)☆19Jan 27, 2026Updated 9 months ago
- Official Implementation of EAGLE-1 (ICML'24), EAGLE-2 (EMNLP'24), and EAGLE-3 (NeurIPS'25).☆2,563Feb 20, 2026Updated 8 months ago