deepseek-ai / LPLBLinks
An early research stage expert-parallel load balancer for MoE models based on linear programming.
☆481Updated last month
Alternatives and similar repositories for LPLB
Users that are interested in LPLB are comparing it to the libraries listed below
Sorting:
- torchcomms: a modern PyTorch communications API☆315Updated last week
- Tile-Based Runtime for Ultra-Low-Latency LLM Inference☆502Updated 2 weeks ago
- ArcticInference: vLLM plugin for high-throughput, low-latency inference☆363Updated last week
- Accelerating MoE with IO and Tile-aware Optimizations☆500Updated 2 weeks ago
- Perplexity open source garden for inference technology☆321Updated 2 weeks ago
- NVIDIA NVSHMEM is a parallel programming interface for NVIDIA GPUs based on OpenSHMEM. NVSHMEM can significantly reduce multi-process com…☆433Updated last week
- ByteCheckpoint: An Unified Checkpointing Library for LFMs☆256Updated last month
- Checkpoint-engine is a simple middleware to update model weights in LLM inference engines☆885Updated this week
- Perplexity GPU Kernels☆548Updated 2 months ago
- Genai-bench is a powerful benchmark tool designed for comprehensive token-level performance evaluation of large language model (LLM) serv…☆250Updated 3 weeks ago
- Autonomous GPU Kernel Generation via Deep Agents☆197Updated last week
- [OSDI'24] Serving LLM-based Applications Efficiently with Semantic Variable☆204Updated last year
- KV cache compression for high-throughput LLM inference☆148Updated 11 months ago
- TritonParse: A Compiler Tracer, Visualizer, and Reproducer for Triton Kernels☆179Updated last week
- dInfer: An Efficient Inference Framework for Diffusion Language Models☆378Updated last week
- Efficient Long-context Language Model Training by Core Attention Disaggregation☆73Updated last week
- ☆81Updated 2 months ago
- ☆654Updated last week
- ☆337Updated this week
- Ship correct and fast LLM kernels to PyTorch☆127Updated 3 weeks ago
- Helpful kernel tutorials and examples for tile-based GPU programming☆526Updated this week
- Easy, Fast, and Scalable Multimodal AI☆83Updated last week
- JAX backend for SGL☆205Updated last week
- Dynamic Memory Management for Serving LLMs without PagedAttention☆454Updated 7 months ago
- Code for data-aware compression of DeepSeek models☆66Updated 3 weeks ago
- ☆116Updated 7 months ago
- [ICLR2025] Breaking Throughput-Latency Trade-off for Long Sequences with Speculative Decoding☆135Updated last year
- ☆96Updated 9 months ago
- Fast low-bit matmul kernels in Triton☆416Updated 3 weeks ago
- PyTorch library for cost-effective, fast and easy serving of MoE models.☆269Updated 2 months ago