deepseek-ai / LPLBLinks
An early research stage expert-parallel load balancer for MoE models based on linear programming.
☆456Updated 3 weeks ago
Alternatives and similar repositories for LPLB
Users that are interested in LPLB are comparing it to the libraries listed below
Sorting:
- ArcticInference: vLLM plugin for high-throughput, low-latency inference☆345Updated this week
- Tile-Based Runtime for Ultra-Low-Latency LLM Inference☆391Updated this week
- torchcomms: a modern PyTorch communications API☆302Updated this week
- Perplexity GPU Kernels☆536Updated last month
- Checkpoint-engine is a simple middleware to update model weights in LLM inference engines☆858Updated this week
- Perplexity open source garden for inference technology☆295Updated this week
- ByteCheckpoint: An Unified Checkpointing Library for LFMs☆256Updated this week
- NVIDIA NVSHMEM is a parallel programming interface for NVIDIA GPUs based on OpenSHMEM. NVSHMEM can significantly reduce multi-process com…☆407Updated last month
- Autonomous GPU Kernel Generation via Deep Agents☆179Updated this week
- Genai-bench is a powerful benchmark tool designed for comprehensive token-level performance evaluation of large language model (LLM) serv…☆236Updated 2 weeks ago
- JAX backend for SGL☆191Updated this week
- ☆564Updated this week
- Code for data-aware compression of DeepSeek models☆66Updated last month
- TritonParse: A Compiler Tracer, Visualizer, and Reproducer for Triton Kernels☆177Updated last week
- KV cache compression for high-throughput LLM inference☆146Updated 10 months ago
- Efficient LLM Inference over Long Sequences☆393Updated 5 months ago
- ☆97Updated 8 months ago
- Helpful kernel tutorials and examples for tile-based GPU programming☆202Updated last week
- Allow torch tensor memory to be released and resumed later☆184Updated last week
- ☆328Updated last month
- Ship correct and fast LLM kernels to PyTorch☆126Updated this week
- ☆114Updated 6 months ago
- Easy, Fast, and Scalable Multimodal AI☆78Updated 2 weeks ago
- [ICLR2025 Spotlight] MagicPIG: LSH Sampling for Efficient LLM Generation☆243Updated 11 months ago
- QeRL enables RL for 32B LLMs on a single H100 GPU.☆466Updated 2 weeks ago
- [ICLR'25] Fast Inference of MoE Models with CPU-GPU Orchestration☆246Updated last year
- PyTorch library for cost-effective, fast and easy serving of MoE models.☆263Updated last month
- Fast low-bit matmul kernels in Triton☆407Updated 3 weeks ago
- [OSDI'24] Serving LLM-based Applications Efficiently with Semantic Variable☆200Updated last year
- kernels, of the mega variety☆623Updated 2 months ago