JetSpec: Breaking the Scaling Ceiling of Speculative Decoding with Causal Parallel Tree Drafting
☆171Jul 31, 2026Updated this week
Alternatives and similar repositories for JetSpec
Users that are interested in JetSpec are comparing it to the libraries listed below. We may earn a commission when you buy through links labeled 'Ad' on this page.
Sorting:
- Official implementation of “Domino: Decoupling Causal Modeling from Autoregressive Drafting in Speculative Decoding”.☆127Jul 25, 2026Updated last week
- [NeurIPS 2025] Scaling Speculative Decoding with Lookahead Reasoning☆69Oct 31, 2025Updated 9 months ago
- [ICML 2026] d3LLM: Ultra-Fast Diffusion LLM 🚀☆149May 1, 2026Updated 3 months ago
- Efficient Long-context Language Model Training by Core Attention Disaggregation☆106Apr 7, 2026Updated 3 months ago
- Train speculative decoding models effortlessly and port them smoothly to SGLang serving.☆1,038Updated this week
- 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.
- DeepSpec: a full-stack codebase for training and evaluating speculative decoding algorithms☆6,865Jul 9, 2026Updated 3 weeks ago
- ☆391Apr 16, 2026Updated 3 months ago
- [ASPLOS'26] Taming the Long-Tail: Efficient Reasoning RL Training with Adaptive Drafter☆176Feb 27, 2026Updated 5 months ago
- Multi-Turn RL Training System with AgentTrainer for Language Model Game Reinforcement Learning☆65Dec 18, 2025Updated 7 months ago
- [NeurIPS 2025] Simple extension on vLLM to help you speed up reasoning model without training.☆232May 31, 2025Updated last year
- Scale digital agent rollouts without pain.☆34Jun 18, 2026Updated last month
- DFlash: Block Diffusion for Flash Speculative Decoding☆5,562May 10, 2026Updated 2 months ago
- Accelerating Large-Scale Reasoning Model Inference with Sparse Self-Speculative Decoding☆117Dec 2, 2025Updated 8 months ago
- An efficient implementation of the NSA (Native Sparse Attention) kernel☆134Jun 24, 2025Updated last year
- Simple, predictable pricing with DigitalOcean hosting • AdAlways know what you'll pay with monthly caps and flat pricing. Enterprise-grade infrastructure trusted by 600k+ customers.
- FlashMemory DS-V4 Retriever: a lightweight retriever that sparsifies DeepSeek-V4 CSA KV-cache. Weights available on Hugging Face.☆99Jul 18, 2026Updated 2 weeks ago
- ☆20Dec 24, 2024Updated last year
- ☆37Aug 7, 2025Updated 11 months ago
- NVFP4 Flash-Attention 4 on BlackWell☆33Jul 23, 2026Updated last week
- high-performance linear attention kernel library built on TileLang☆625Jul 28, 2026Updated last week
- [ICML 2026] Jacobi Forcing: Fast and Accurate Diffusion-style Decoding☆125Feb 20, 2026Updated 5 months ago
- Vortex: Programmable Sparse Attention for Agents as Algorithm Designers☆68Updated this week
- Official Implementation of SAM-Decoding: Speculative Decoding via Suffix Automaton☆52May 12, 2026Updated 2 months ago
- Compact and Agent-Native MoE Training System☆314Updated this week
- Virtual machines for every use case on DigitalOcean • AdGet dependable uptime with 99.99% SLA, simple security tools, and predictable monthly pricing with DigitalOcean's virtual machines, called Droplets.
- A Survey of Efficient Attention Methods: Hardware-efficient, Sparse, Compact, and Linear Attention☆305Dec 1, 2025Updated 8 months ago
- TokenSpeed is a speed-of-light LLM inference engine.☆1,795Updated this week
- An experimental communicating attention kernel based on DeepEP.☆34Jul 29, 2025Updated last year
- Memory-bounded compressed sparse attention via streaming top-k. Triton kernels for the DeepSeek-V4 lightning indexer. 32x regime extensio…☆23May 5, 2026Updated 2 months ago
- ☆47Jul 27, 2026Updated last week
- ☆136May 29, 2025Updated last year
- Kinetics: Rethinking Test-Time Scaling Laws☆87Jul 11, 2025Updated last year
- FlexAttention w/ FlashAttention3 Support☆27Oct 5, 2024Updated last year
- ☆53May 19, 2025Updated last year
- GPU virtual machines on DigitalOcean Gradient AI • AdGet to production fast with high-performance AMD and NVIDIA GPUs you can spin up in seconds. The definition of operational simplicity.
- Nano SGLang☆16Jul 21, 2025Updated last year
- ☆16Oct 11, 2025Updated 9 months ago
- An agent for CUDA compute-communication kernel co-design☆36May 7, 2026Updated 2 months ago
- ArcticInference: vLLM plugin for high-throughput, low-latency inference☆463Jul 14, 2026Updated 2 weeks ago
- ☆46Oct 15, 2025Updated 9 months ago
- [KernelGYM & Dr. Kernel] A distributed GPU environment and a collection of RL training methods to support RL for Kernel Generations [ICML…☆200Mar 29, 2026Updated 4 months ago
- Official Implementation of EAGLE-1 (ICML'24), EAGLE-2 (EMNLP'24), and EAGLE-3 (NeurIPS'25).☆2,494Feb 20, 2026Updated 5 months ago