DeepSpec: a full-stack codebase for training and evaluating speculative decoding algorithms
☆6,993Jul 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☆5,643May 10, 2026Updated 3 months ago
- Train speculative decoding models effortlessly and port them smoothly to SGLang serving.☆1,090Updated this week
- Official implementation of “Domino: Decoupling Causal Modeling from Autoregressive Drafting in Speculative Decoding”.☆132Jul 25, 2026Updated 3 weeks ago
- slime is an LLM post-training framework for RL Scaling.☆8,119Updated this week
- A kernel library written in tilelang☆1,733Apr 23, 2026Updated 3 months 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.
- TokenSpeed is a speed-of-light LLM inference engine.☆1,926Updated this week
- Domain-specific language designed to streamline the development of high-performance GPU/CPU/Accelerators kernels☆7,241Updated this week
- JetSpec: Breaking the Scaling Ceiling of Speculative Decoding with Causal Parallel Tree Drafting☆176Aug 9, 2026Updated last week
- SGLang is a high-performance serving framework for large language models and multimodal models.☆32,029Updated this week
- FlashKDA: high-performance Kimi Delta Attention kernels☆1,214Jul 30, 2026Updated 2 weeks ago
- FlashInfer: Kernel Library for LLM Serving☆6,187Updated this week
- DeepGEMM: clean and efficient BLAS kernel library on GPU☆7,700Aug 11, 2026Updated last week
- Mooncake is the serving platform for Kimi, a leading LLM service provided by Moonshot AI.☆6,313Updated this week
- high-performance linear attention kernel library built on TileLang☆641Aug 11, 2026Updated last week
- Open source password manager - Proton Pass • AdSecurely store, share, and autofill your credentials with Proton Pass, the end-to-end encrypted password manager trusted by millions.
- Conditional Memory via Scalable Lookup: A New Axis of Sparsity for Large Language Models☆4,596Jan 14, 2026Updated 7 months ago
- Tile-Based Runtime for Ultra-Low-Latency LLM Inference☆1,708Updated this week
- A high-throughput and memory-efficient inference and serving engine for LLMs☆89,360Updated this week
- LMCache: Supercharge Your LLM with the Fastest KV Cache Layer☆11,199Updated this week
- Accelerating MoE with IO and Tile-aware Optimizations☆745Updated this week
- Nano vLLM☆15,046Apr 26, 2026Updated 3 months ago
- ☆411Jul 30, 2026Updated 2 weeks ago
- A unified library for building, evaluating, and storing speculative decoding algorithms for LLM inference in vLLM☆739Updated this week
- verl/HybridFlow: A Flexible and Efficient RL Post-Training Framework☆23,022Updated this week
- 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.
- A compact implementation of SGLang, designed to demystify the complexities of modern LLM serving systems.☆4,767May 17, 2026Updated 3 months ago
- 🚀 Efficient implementations for emerging model architectures☆5,575Updated this week
- Distributed Compiler and Optimized Parallel Kernels☆1,518Aug 12, 2026Updated last week
- Official Implementation of EAGLE-1 (ICML'24), EAGLE-2 (EMNLP'24), and EAGLE-3 (NeurIPS'25).☆2,502Feb 20, 2026Updated 5 months ago
- Miles is an enterprise-facing reinforcement learning framework for LLM and VLM post-training, forked from and co-evolving with slime.☆2,028Updated this week
- High Performance LLM Inference Operator Library☆1,119Aug 6, 2026Updated last week
- A framework for efficient model inference with omni-modality models☆6,150Updated this week
- A Quirky Assortment of CuTe Kernels☆1,117Updated this week
- A PyTorch native library for training speculative decoding models☆226Updated this week
- Deploy to Railway using AI coding agents - Free Credits Offer • AdUse Claude Code, Codex, OpenCode, and more. Autonomous software development now has the infrastructure to match with Railway.
- DeepSeek 4 Flash and PRO local inference engine for Metal, CUDA and ROCm☆21,528Aug 9, 2026Updated last week
- ☆840Jun 2, 2026Updated 2 months ago
- Run frontier MoE models on hardware you already own — pure C, zero deps, experts streamed from disk. Tiny engine, immense model. 🐦☆25,406Updated this week
- A Flexible Framework for Experiencing Heterogeneous LLM Inference/Fine-tune Optimizations☆19,257Updated this week
- mKernel: fast multi-node, multi-GPU fused kernels☆266Updated this week
- A fast communication-overlapping library for tensor/expert parallelism on GPUs.☆1,352Aug 28, 2025Updated 11 months ago
- Local UI to run and train LLMs and diffusion models, including Qwen3.8, Kimi K3, MiniMax-H3, Gemma 4, DeepSeek-V4, FLUX and more.☆73,589Updated this week