The official implementation for the intra-stage fusion technique introduced in https://arxiv.org/abs/2409.13221
☆31Apr 22, 2025Updated last year
Alternatives and similar repositories for FlexFusion
Users that are interested in FlexFusion are comparing it to the libraries listed below. We may earn a commission when you buy through links labeled 'Ad' on this page.
Sorting:
- To pioneer training long-context multi-modal transformer models☆75Aug 8, 2025Updated 11 months ago
- ☆27Oct 1, 2025Updated 9 months ago
- Sample Codes using NVSHMEM on Multi-GPU☆30Jan 22, 2023Updated 3 years ago
- ☆45Oct 15, 2025Updated 9 months ago
- ☆83Feb 5, 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.
- [Archived] For the latest updates and community contribution, please visit: https://github.com/Ascend/TransferQueue or https://gitcode.co…☆16Jan 16, 2026Updated 6 months ago
- An experimental communicating attention kernel based on DeepEP.☆34Jul 29, 2025Updated 11 months ago
- Arcaea魔改版XD 合成大風暴☆18Feb 1, 2021Updated 5 years ago
- A NCCL extension library, designed to efficiently offload GPU memory allocated by the NCCL communication library.☆110Dec 17, 2025Updated 7 months ago
- SC'25 UltraAttn: Efficiently Parallelizing Attention through Hierarchical Context-Tiling☆16Aug 14, 2025Updated 11 months ago
- ☆18May 28, 2024Updated 2 years ago
- [ICML 2026] "LIVE: Long-horizon Interactive Video World ModEling"☆35Jul 15, 2026Updated last week
- NCCL communication API layer, and transport layer created from first principles.☆16Aug 20, 2025Updated 11 months ago
- ☆56Apr 30, 2025Updated last year
- 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.
- ☆19May 9, 2025Updated last year
- Proposal for the next generation of course-oriented IR.☆10Dec 24, 2021Updated 4 years ago
- Code repository for the SOSP'25 paper DCP: Addressing Input Dynamism In Long-Context Training via Dynamic Context Parallelism.☆21Nov 28, 2025Updated 7 months ago
- Communication patterns for AI, built on top of NCCL device and host APIs☆18Updated this week
- ☆37Updated this week
- Allow torch tensor memory to be released and resumed later☆260Updated this week
- [Experimental] Miles-diffusion is an post-training framework for large-scale diffusion model training and production workloads, forked fr…☆22Updated this week
- A Triton-only attention backend for vLLM☆27Jul 14, 2026Updated last week
- Nex Venus Communication Library☆75Nov 17, 2025Updated 8 months ago
- 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.
- Compositional Muon release☆23Jun 5, 2026Updated last month
- Paper reading and discussion notes, covering AI frameworks, distributed systems, cluster management, etc.☆69Mar 4, 2026Updated 4 months ago
- Web front for EESAST☆11Updated this week
- Prototyp MegaScale-Infer: Serving Mixture-of-Experts at Scale with Disaggregated Expert Parallelism☆31Apr 4, 2025Updated last year
- ☆13Jun 14, 2026Updated last month
- patches for huggingface transformers to save memory☆36May 9, 2026Updated 2 months ago
- ☆26Feb 17, 2025Updated last year
- DeepXTrace is a lightweight tool for precisely diagnosing slow ranks in DeepEP-based environments.☆101Jan 16, 2026Updated 6 months ago
- NexRL is an ultra-loosely-coupled LLM post-training framework.☆114May 13, 2026Updated 2 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.
- Dynamic resources changes for multi-dimensional parallelism training☆31Aug 22, 2025Updated 11 months ago
- DeepGEMM: clean and efficient FP8 GEMM kernels with fine-grained scaling☆32Updated this week
- Tempo is a system for declarative, efficient, end-to-end compiled dynamic deep learning☆30Oct 21, 2025Updated 9 months ago
- Bridge Megatron-Core to Hugging Face/Reinforcement Learning☆226Jun 15, 2026Updated last month
- Benchmark workloads of Boki☆11Sep 8, 2021Updated 4 years ago
- The Official Repository for Paper "HarmfulSkillBench: How Do Harmful Skills Weaponize Your Agents?"☆15May 2, 2026Updated 2 months ago
- 研究生课《网络大数据管理理论和应用》大作业项目代码☆13Dec 31, 2022Updated 3 years ago