An early research stage expert-parallel load balancer for MoE models based on linear programming.
☆530Nov 19, 2025Updated 9 months ago
Alternatives and similar repositories for LPLB
Users that are interested in LPLB are comparing it to the libraries listed below. We may earn a commission when you buy through links labeled 'Ad' on this page.
Sorting:
- ☆46Oct 15, 2025Updated 10 months ago
- Accelerating MoE with IO and Tile-aware Optimizations☆750Updated this week
- ☆384Jan 28, 2026Updated 7 months ago
- Distributed Compiler and Optimized Parallel Kernels☆1,529Aug 12, 2026Updated 2 weeks ago
- Perplexity GPU Kernels☆604Nov 7, 2025Updated 9 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.
- An experimental communicating attention kernel based on DeepEP.☆34Jul 29, 2025Updated last year
- DeepSeek-V3.2-Exp DSA Warmup Lightning Indexer training operator based on tilelang☆52Nov 19, 2025Updated 9 months ago
- A fast communication-overlapping library for tensor/expert parallelism on GPUs.☆1,355Aug 28, 2025Updated last year
- ☆32Jul 2, 2025Updated last year
- DeeperGEMM: crazy optimized version☆85May 5, 2025Updated last year
- High-performance LLM operator library built on TileLang.☆173Updated this week
- Tile-based language built for AI computation across all scales☆188Updated this week
- mKernel: fast multi-node, multi-GPU fused kernels☆270Updated this week
- Nex Venus Communication Library☆75Nov 17, 2025Updated 9 months ago
- 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.
- Checkpoint-engine is a simple middleware to update model weights in LLM inference engines☆1,003Aug 12, 2026Updated 2 weeks ago
- NVSHMEM‑Tutorial: Build a DeepEP‑like GPU Buffer☆200Feb 11, 2026Updated 6 months ago
- NVIDIA NVSHMEM is a parallel programming interface for NVIDIA GPUs based on OpenSHMEM. NVSHMEM can significantly reduce multi-process com…☆578Updated this week
- A Quirky Assortment of CuTe Kernels☆1,134Aug 21, 2026Updated last week
- A kernel library written in tilelang☆1,743Apr 23, 2026Updated 4 months ago
- CUDA kernels for linear attention variants, written in CuTe DSL and CUTLASS C++.☆543Updated this week
- ☆253Nov 19, 2025Updated 9 months ago
- Accepted to MLSys 2026☆94Apr 19, 2026Updated 4 months ago
- Low overhead tracing library and trace visualizer for pipelined CUDA kernels☆135Jul 29, 2026Updated last month
- 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.
- FlashKDA: high-performance Kimi Delta Attention kernels☆1,236Jul 30, 2026Updated last month
- Efficient Long-context Language Model Training by Core Attention Disaggregation☆105Apr 7, 2026Updated 4 months ago
- Distributed MoE in a Single Kernel [NeurIPS '25]☆288May 5, 2026Updated 3 months ago
- DeepGEMM: clean and efficient BLAS kernel library on GPU☆7,746Updated this week
- Utility scripts for PyTorch (e.g. Make Perfetto show some disappearing kernels, Memory profiler that understands more low-level allocatio…☆116Sep 11, 2025Updated 11 months ago
- A lightweight design for computation-communication overlap.☆245Jan 20, 2026Updated 7 months ago
- A NCCL extension library, designed to efficiently offload GPU memory allocated by the NCCL communication library.☆116Dec 17, 2025Updated 8 months ago
- ☆37Aug 7, 2025Updated last year
- Learning TileLang with 10 puzzles!☆373Aug 11, 2026Updated 2 weeks ago
- 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.
- Tile-Based Runtime for Ultra-Low-Latency LLM Inference☆1,738Aug 13, 2026Updated 2 weeks ago
- torchcomms: a modern PyTorch communications API☆390Updated this week
- 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
- Expert Parallelism Load Balancer☆1,425Mar 24, 2025Updated last year
- NVIDIA cuTile learn☆168Dec 9, 2025Updated 8 months ago
- Compact and Agent-Native MoE Training System☆344Aug 22, 2026Updated last week
- Domain-specific language designed to streamline the development of high-performance GPU/CPU/Accelerators kernels☆7,302Updated this week