DeepSpec: a full-stack codebase for training and evaluating speculative decoding algorithms
☆7,086Jul 9, 2026Updated last month
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☆6,061Aug 18, 2026Updated 3 weeks ago
- Train speculative decoding models effortlessly and port them smoothly to SGLang serving.☆1,152Updated this week
- slime is an LLM post-training framework for RL Scaling.☆8,411Updated this week
- [EMNLP2026 Main] Official implementation of “Domino: Decoupling Causal Modeling from Autoregressive Drafting in Speculative Decoding”.☆138Jul 25, 2026Updated last month
- A kernel library written in tilelang☆1,755Apr 23, 2026Updated 4 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.☆2,104Updated this week
- Domain-specific language designed to streamline the development of high-performance GPU/CPU/Accelerators kernels☆7,368Updated this week
- JetSpec: Breaking the Scaling Ceiling of Speculative Decoding with Causal Parallel Tree Drafting☆218Aug 9, 2026Updated 3 weeks ago
- SGLang is a high-performance serving framework for large language models and multimodal models.☆35,647Updated this week
- FlashKDA: high-performance Kimi Delta Attention kernels☆1,247Sep 1, 2026Updated last week
- FlashInfer: Kernel Library for LLM Serving☆6,354Updated this week
- DeepGEMM: clean and efficient BLAS kernel library on GPU☆7,789Updated this week
- Mooncake is the serving platform for Kimi, a leading LLM service provided by Moonshot AI.☆6,526Updated this week
- high-performance linear attention kernel library built on TileLang☆691Aug 26, 2026Updated last week
- Managed Kubernetes at scale on DigitalOcean • AdDigitalOcean Kubernetes includes the control plane, bandwidth allowance, container registry, automatic updates, and more for free.
- Tile-Based Runtime for Ultra-Low-Latency LLM Inference☆1,758Aug 13, 2026Updated 3 weeks ago
- Conditional Memory via Scalable Lookup: A New Axis of Sparsity for Large Language Models☆4,642Jan 14, 2026Updated 7 months ago
- A high-throughput and memory-efficient inference and serving engine for LLMs☆91,254Updated this week
- LMCache: Supercharge Your LLM with the Fastest KV Cache Layer☆11,704Updated this week
- Accelerating MoE with IO and Tile-aware Optimizations☆761Aug 29, 2026Updated last week
- Nano vLLM☆15,364Apr 26, 2026Updated 4 months ago
- ☆420Jul 30, 2026Updated last month
- A unified library for building, evaluating, and storing speculative decoding algorithms for LLM inference in vLLM☆820Updated this week
- verl/HybridFlow: A Flexible and Efficient RL Post-Training Framework☆23,344Updated this week
- Wordpress hosting with auto-scaling - Free Trial Offer • AdFully Managed hosting for WordPress and WooCommerce businesses that need reliable, auto-scalable performance. Cloudways SafeUpdates now available.
- A compact implementation of SGLang, designed to demystify the complexities of modern LLM serving systems.☆5,008May 17, 2026Updated 3 months ago
- 🚀 Efficient implementations for emerging model architectures☆5,717Updated this week
- Distributed Compiler and Optimized Parallel Kernels☆1,540Aug 12, 2026Updated 3 weeks ago
- Official Implementation of EAGLE-1 (ICML'24), EAGLE-2 (EMNLP'24), and EAGLE-3 (NeurIPS'25).☆2,527Feb 20, 2026Updated 6 months ago
- Miles is an enterprise-facing reinforcement learning framework for LLM and VLM post-training, forked from and co-evolving with slime.☆2,668Updated this week
- High Performance LLM Inference Operator Library☆1,150Updated this week
- A framework for efficient model inference with omni-modality models☆6,719Updated this week
- A Quirky Assortment of CuTe Kernels☆1,138Updated this week
- A PyTorch native library for training speculative decoding models☆244Updated this week
- 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.
- DeepSeek 4 Flash and PRO local inference engine for Metal, CUDA and ROCm☆22,161Updated this week
- ☆961Sep 2, 2026Updated last week
- Run frontier MoE models on hardware you already own — pure C, zero deps, experts streamed from disk. Tiny engine, immense model. 🐦☆27,002Updated this week
- A Flexible Framework for Experiencing Heterogeneous LLM Inference/Fine-tune Optimizations☆19,480Updated this week
- mKernel: fast multi-node, multi-GPU fused kernels☆270Updated this week
- A fast communication-overlapping library for tensor/expert parallelism on GPUs.☆1,359Aug 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.☆75,870Updated this week