tile-ai / TileRTLinks
Tile-Based Runtime for Ultra-Low-Latency LLM Inference
☆557Updated this week
Alternatives and similar repositories for TileRT
Users that are interested in TileRT are comparing it to the libraries listed below
Sorting:
- An early research stage expert-parallel load balancer for MoE models based on linear programming.☆491Updated 2 months ago
- Perplexity GPU Kernels☆554Updated 2 months ago
- A lightweight design for computation-communication overlap.☆213Updated last week
- NVIDIA NVSHMEM is a parallel programming interface for NVIDIA GPUs based on OpenSHMEM. NVSHMEM can significantly reduce multi-process com…☆459Updated 3 weeks ago
- Open ABI and FFI for Machine Learning Systems☆307Updated last week
- Allow torch tensor memory to be released and resumed later☆207Updated 2 weeks ago
- ☆340Updated 3 weeks ago
- Tile-based language built for AI computation across all scales☆117Updated 2 weeks ago
- Accelerating MoE with IO and Tile-aware Optimizations☆563Updated last week
- Dynamic Memory Management for Serving LLMs without PagedAttention☆457Updated 8 months ago
- High Performance LLM Inference Operator Library☆222Updated last week
- Genai-bench is a powerful benchmark tool designed for comprehensive token-level performance evaluation of large language model (LLM) serv…☆252Updated last week
- torchcomms: a modern PyTorch communications API☆323Updated this week
- Perplexity open source garden for inference technology☆350Updated last month
- Train speculative decoding models effortlessly and port them smoothly to SGLang serving.☆659Updated this week
- Autonomous GPU Kernel Generation via Deep Agents☆223Updated this week
- High performance Transformer implementation in C++.☆148Updated last year
- A tiny yet powerful LLM inference system tailored for researching purpose. vLLM-equivalent performance with only 2k lines of code (2% of …☆309Updated 7 months ago
- DeepSeek-V3/R1 inference performance simulator☆177Updated 10 months ago
- NVSHMEM‑Tutorial: Build a DeepEP‑like GPU Buffer☆154Updated 4 months ago
- nnScaler: Compiling DNN models for Parallel Training☆124Updated 4 months ago
- NVIDIA cuTile learn☆150Updated last month
- ArcticInference: vLLM plugin for high-throughput, low-latency inference☆379Updated last week
- A low-latency & high-throughput serving engine for LLMs☆470Updated 3 weeks ago
- High-performance distributed data shuffling (all-to-all) library for MoE training and inference☆109Updated 3 weeks ago
- ☆96Updated 10 months ago
- Building the Virtuous Cycle for AI-driven LLM Systems☆140Updated this week
- ☆93Updated 9 months ago
- 🤖FFPA: Extend FlashAttention-2 with Split-D, ~O(1) SRAM complexity for large headdim, 1.8x~3x↑🎉 vs SDPA EA.☆246Updated last week
- PyTorch library for cost-effective, fast and easy serving of MoE models.☆276Updated 3 months ago