DeepSpec: a full-stack codebase for training and evaluating speculative decoding algorithms
☆7,295Jul 9, 2026Updated 4 months ago
Alternatives and similar repositories for DeepSpec
Users that are interested in DeepSpec are comparing it to the libraries listed below. We may earn a commission when you buy through links labeled 'Ad' on this page.
Sorting:
- Train speculative decoding models effortlessly and port them smoothly to SGLang serving.☆1,273Updated this week
- DFlash: Block Diffusion for Flash Speculative Decoding☆6,220Aug 18, 2026Updated 2 months ago
- slime is an LLM training framework for RL Scaling.☆8,807Updated this week
- DeepInfer: High-performance, large-scale inference framework for DeepSeek models☆25,655Updated this week
- A kernel library written in tilelang☆2,035Oct 14, 2026Updated 3 weeks ago
- 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.
- MoonEP: A Perfectly Balanced Expert Parallelism Library via Dynamic Redundant Experts☆1,203Oct 12, 2026Updated 3 weeks ago
- A unified library for building, evaluating, and storing speculative decoding algorithms for LLM inference in vLLM☆945Updated this week
- SGLang is a high-performance serving framework for large language models and multimodal models.☆38,577Updated this week
- TokenSpeed is a speed-of-light LLM inference engine.☆2,888Updated this week
- Domain-specific language designed to streamline the development of high-performance GPU/CPU/Accelerators kernels☆8,929Updated this week
- [EMNLP2026 Main] Official implementation of “Domino: Decoupling Causal Modeling from Autoregressive Drafting in Speculative Decoding”.☆141Jul 25, 2026Updated 3 months ago
- DeepGEMM: clean and efficient BLAS kernel library on GPU☆9,171Oct 20, 2026Updated 2 weeks ago
- FlashInfer: Kernel Library for LLM Serving☆6,875Updated this week
- FlashKDA: high-performance Kimi Delta Attention kernels☆1,296Oct 26, 2026Updated 2 weeks ago
- AI Agents on DigitalOcean Gradient AI Platform • AdBuild production-ready AI agents using customizable tools or access multiple LLMs through a single endpoint. Create custom knowledge bases or connect external data.
- Mooncake is the serving platform for Kimi, a leading LLM service provided by Moonshot AI.☆7,003Updated this week
- DeepSelect: TopK kernels for DeepSeek Sparse Attention (DSA) and Samplers☆533Sep 30, 2026Updated last month
- Tile-Based Runtime for Ultra-Low-Latency LLM Inference☆1,887Sep 24, 2026Updated last month
- high-performance linear attention kernel library built on TileLang☆750Oct 30, 2026Updated last week
- Conditional Memory via Scalable Lookup: A New Axis of Sparsity for Large Language Models☆4,815Jan 14, 2026Updated 9 months ago
- DeepEP: an efficient expert-parallel communication library☆10,420Oct 21, 2026Updated 2 weeks ago
- Miles is an enterprise-facing reinforcement learning framework for LLM and VLM post-training, forked from and co-evolving with slime.☆3,279Updated this week
- JetSpec++/JetSpec: Causal Parallel (Tree) Drafting with Self Refinement☆256Updated this week
- LMCache: Supercharge Your LLM with the Fastest KV Cache Layer☆12,310Updated 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.
- AgentENV (AENV) is a distributed platform for running agent environments at scale.☆3,732Updated this week
- A high-throughput and memory-efficient inference and serving engine for LLMs☆96,551Updated this week
- 🚀 Efficient implementations for emerging model architectures☆6,027Updated this week
- FlashMLA: Efficient Multi-head Latent Attention Kernels☆13,255Oct 27, 2026Updated last week
- Kernel Design Agents (KDA) is a agent-centric workflow to write high-performance CUDA Kernels.☆1,550Oct 19, 2026Updated 3 weeks ago
- Nano vLLM☆16,360Apr 26, 2026Updated 6 months ago
- Accelerating MoE with IO and Tile-aware Optimizations☆787Aug 29, 2026Updated 2 months ago
- verl/HybridFlow: A Flexible and Efficient RL Post-Training Framework☆24,165Updated this week
- A compact implementation of SGLang, designed to demystify the complexities of modern LLM serving systems.☆5,539May 17, 2026Updated 5 months ago
- 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 PyTorch native library for training speculative decoding models☆267Updated this week
- Official Implementation of EAGLE-1 (ICML'24), EAGLE-2 (EMNLP'24), and EAGLE-3 (NeurIPS'25).☆2,574Feb 20, 2026Updated 8 months ago
- A lightweight library for xPU kernel JIT compilation☆456Updated this week
- Ongoing research training transformer models at scale☆18,349Updated this week
- Fast and memory-efficient exact attention☆25,257Updated this week
- High Performance LLM Inference Operator Library☆1,190Sep 7, 2026Updated 2 months ago
- Distributed Compiler and Optimized Parallel Kernels☆1,577Sep 18, 2026Updated last month