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☆76Aug 8, 2025Updated last year
- ☆27Oct 1, 2025Updated 10 months ago
- Sample Codes using NVSHMEM on Multi-GPU☆30Jan 22, 2023Updated 3 years ago
- ☆46Oct 15, 2025Updated 10 months ago
- ☆92Feb 5, 2026Updated 6 months ago
- Proton VPN Special Offer - Get 70% off • AdSpecial partner offer. Trusted by over 100 million users worldwide. Tested, Approved and Recommended by Experts.
- [Archived] For the latest updates and community contribution, please visit: https://github.com/Ascend/TransferQueue or https://gitcode.co…☆15Aug 14, 2026Updated 2 weeks ago
- An experimental communicating attention kernel based on DeepEP.☆34Jul 29, 2025Updated last year
- WHU Academic Calendar☆13Updated this week
- 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.☆116Dec 17, 2025Updated 8 months ago
- SC'25 UltraAttn: Efficiently Parallelizing Attention through Hierarchical Context-Tiling☆16Aug 14, 2025Updated last year
- ☆18May 28, 2024Updated 2 years ago
- [ICML 2026] "LIVE: Long-horizon Interactive Video World ModEling"☆43Jul 15, 2026Updated last month
- NCCL communication API layer, and transport layer created from first principles.☆16Aug 20, 2025Updated last year
- 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.
- ☆56Apr 30, 2025Updated last year
- ☆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.☆22Nov 28, 2025Updated 9 months ago
- Communication patterns for AI, built on top of NCCL device and host APIs☆42Updated this week
- ☆37Updated this week
- Allow torch tensor memory to be released and resumed later☆272Aug 21, 2026Updated last week
- Miles-diffusion is an post-training framework for large-scale diffusion model training and production workloads, forked from and co-evolv…☆70Updated this week
- A Triton-only attention backend for vLLM☆28Jul 14, 2026Updated last month
- 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.
- Nex Venus Communication Library☆75Nov 17, 2025Updated 9 months ago
- Paper reading and discussion notes, covering AI frameworks, distributed systems, cluster management, etc.☆73Mar 4, 2026Updated 5 months ago
- Compositional Muon release☆25Jun 5, 2026Updated 2 months ago
- Web front for EESAST☆12Updated this week
- Prototyp MegaScale-Infer: Serving Mixture-of-Experts at Scale with Disaggregated Expert Parallelism☆34Apr 4, 2025Updated last year
- ☆13Aug 14, 2026Updated 2 weeks ago
- patches for huggingface transformers to save memory☆37May 9, 2026Updated 3 months ago
- ☆27Feb 17, 2025Updated last year
- DeepXTrace is a lightweight tool for precisely diagnosing slow ranks in DeepEP-based environments.☆102Jan 16, 2026Updated 7 months ago
- GPUs on demand by Runpod - Special Offer Available • AdRun AI, ML, and HPC workloads on powerful cloud GPUs—without limits or wasted spend. Deploy GPUs in under a minute and pay by the second.
- NexRL is an ultra-loosely-coupled LLM post-training framework.☆119Jul 30, 2026Updated last month
- Dynamic resources changes for multi-dimensional parallelism training☆31Aug 22, 2025Updated last year
- DeepGEMM: clean and efficient FP8 GEMM kernels with fine-grained scaling☆38Updated this week
- Tempo is a system for declarative, efficient, end-to-end compiled dynamic deep learning☆31Oct 21, 2025Updated 10 months ago
- All you need to get started with the LM Playpen Environment for Learning in Interaction.☆17Aug 20, 2026Updated last week
- Bridge Megatron-Core to Hugging Face/Reinforcement Learning☆231Jun 15, 2026Updated 2 months ago
- Benchmark workloads of Boki☆11Sep 8, 2021Updated 4 years ago