DeepSpec: a full-stack codebase for training and evaluating speculative decoding algorithms
☆7,167Jul 9, 2026Updated 2 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:
- DFlash: Block Diffusion for Flash Speculative Decoding☆6,123Aug 18, 2026Updated last month
- Train speculative decoding models effortlessly and port them smoothly to SGLang serving.☆1,194Updated this week
- slime is an LLM post-training framework for RL Scaling.☆8,547Updated this week
- [EMNLP2026 Main] Official implementation of “Domino: Decoupling Causal Modeling from Autoregressive Drafting in Speculative Decoding”.☆138Jul 25, 2026Updated 2 months ago
- A kernel library written in tilelang☆1,777Apr 23, 2026Updated 5 months ago
- 1-Click AI Models by DigitalOcean Gradient • AdDeploy popular AI models on DigitalOcean Gradient GPU virtual machines with just a single click. Zero configuration with optimized deployments.
- TokenSpeed is a speed-of-light LLM inference engine.☆2,180Updated this week
- Domain-specific language designed to streamline the development of high-performance GPU/CPU/Accelerators kernels☆7,487Updated this week
- SGLang is a high-performance serving framework for large language models and multimodal models.☆36,481Updated this week
- JetSpec: Breaking the Scaling Ceiling of Speculative Decoding with Causal Parallel Tree Drafting☆226Sep 21, 2026Updated last week
- FlashInfer: Kernel Library for LLM Serving☆6,510Updated this week
- FlashKDA: high-performance Kimi Delta Attention kernels☆1,263Sep 1, 2026Updated 3 weeks ago
- DeepGEMM: clean and efficient BLAS kernel library on GPU☆7,875Updated this week
- Mooncake is the serving platform for Kimi, a leading LLM service provided by Moonshot AI.☆6,668Updated this week
- high-performance linear attention kernel library built on TileLang☆710Updated this week
- Deploy open-source AI quickly and easily - Special Bonus Offer • AdRunpod Hub is built for open source. One-click deployment and autoscaling endpoints without provisioning your own infrastructure.
- Conditional Memory via Scalable Lookup: A New Axis of Sparsity for Large Language Models☆4,701Jan 14, 2026Updated 8 months ago
- Tile-Based Runtime for Ultra-Low-Latency LLM Inference☆1,813Updated this week
- LMCache: Supercharge Your LLM with the Fastest KV Cache Layer☆11,922Updated this week
- A high-throughput and memory-efficient inference and serving engine for LLMs☆92,771Updated this week
- Accelerating MoE with IO and Tile-aware Optimizations☆771Aug 29, 2026Updated last month
- Nano vLLM☆15,636Apr 26, 2026Updated 5 months ago
- A unified library for building, evaluating, and storing speculative decoding algorithms for LLM inference in vLLM☆854Updated this week
- ☆424Sep 19, 2026Updated last week
- verl/HybridFlow: A Flexible and Efficient RL Post-Training Framework☆23,653Updated 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.
- 🚀 Efficient implementations for emerging model architectures☆5,795Updated this week
- A compact implementation of SGLang, designed to demystify the complexities of modern LLM serving systems.☆5,182May 17, 2026Updated 4 months ago
- Distributed Compiler and Optimized Parallel Kernels☆1,552Sep 18, 2026Updated last week
- Official Implementation of EAGLE-1 (ICML'24), EAGLE-2 (EMNLP'24), and EAGLE-3 (NeurIPS'25).☆2,542Feb 20, 2026Updated 7 months ago
- Miles is an enterprise-facing reinforcement learning framework for LLM and VLM post-training, forked from and co-evolving with slime.☆3,008Updated this week
- High Performance LLM Inference Operator Library☆1,168Sep 7, 2026Updated 3 weeks ago
- A framework for efficient model inference with omni-modality models☆7,084Updated this week
- A Quirky Assortment of CuTe Kernels☆1,157Sep 17, 2026Updated last week
- A PyTorch native library for training speculative decoding models☆250Updated this week
- 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.
- DeepSeek 4 Flash and PRO local inference engine for Metal, CUDA and ROCm☆22,736Sep 20, 2026Updated last week
- Kernel Design Agents (KDA) is a agent-centric workflow to write high-performance CUDA Kernels.☆1,108Sep 14, 2026Updated 2 weeks ago
- A Flexible Framework for Experiencing Heterogeneous LLM Inference/Fine-tune Optimizations☆19,539Updated this week
- Run frontier MoE models on hardware you already own — pure C, zero deps, experts streamed from disk. Tiny engine, immense model. 🐦☆37,896Updated this week
- mKernel: fast multi-node, multi-GPU fused kernels☆282Sep 20, 2026Updated last week
- A fast communication-overlapping library for tensor/expert parallelism on GPUs.☆1,367Aug 28, 2025Updated last year
- Local UI to run and train LLMs and diffusion models. Supports GGUF, MLX, Qwen3.8, DeepSeek-V4, MiniMax-H3, Gemma 4, FLUX and more.☆76,860Updated this week