Aggressive decode optimizations for Qwen3-0.6B on RTX 5090
☆64Feb 25, 2026Updated 7 months ago
Alternatives and similar repositories for qwen_megakernel
Users that are interested in qwen_megakernel are comparing it to the libraries listed below. We may earn a commission when you buy through links labeled 'Ad' on this page.
Sorting:
- Qwen3-0.6B megakernel: 527 tok/s decode on RTX 3090 (3.8x faster than PyTorch)☆141Feb 10, 2026Updated 8 months ago
- Region-level profiling for CUDA kernels with trace, NVBit, CUPTI, NSys, and an interactive Explorer.☆133Apr 17, 2026Updated 5 months ago
- Super fast serving stack for LLM on Windows/Linux/Macos☆17Sep 18, 2026Updated 3 weeks ago
- MetaAttention: A Unified and Performant Attention Framework Across Hardware Backends(PPoPP'26)☆18Aug 6, 2026Updated 2 months ago
- High-performance GPU kernels for LLM inference in OpenAI Triton. Fused RMSNorm, SwiGLU, INT8 GEMM with benchmarks and roofline analysis.☆45Jul 22, 2026Updated 2 months ago
- Serverless GPU API endpoints on Runpod - Get Bonus Credits • AdSkip the infrastructure headaches. Auto-scaling, pay-as-you-go, no-ops approach lets you focus on innovating your application.
- mKernel: fast multi-node, multi-GPU fused kernels☆289Updated this week
- ☆18Dec 2, 2024Updated last year
- Implementation of a Light Recurrent Unit in Pytorch☆50Oct 6, 2024Updated 2 years ago
- FlexTensor is a tensor offloading and management library for PyTorch that enables running large models on limited GPU memory by intellige…☆116Aug 18, 2026Updated last month
- mHC kernels implemented in CUDA☆267Mar 9, 2026Updated 7 months ago
- Unified tracing profiler and visualizer, one timeline from CPU to GPU/HES to in-kernel zones☆148Sep 21, 2026Updated 2 weeks ago
- A NCCL extension library, designed to efficiently offload GPU memory allocated by the NCCL communication library.☆119Dec 17, 2025Updated 9 months ago
- ☆19Jan 4, 2024Updated 2 years ago
- Efficient Long-context Language Model Training by Core Attention Disaggregation☆105Apr 7, 2026Updated 6 months ago
- 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.
- Row-wise block scaling for fp8 quantization matrix multiplication. Solution to GPU mode AMD challenge.☆19Feb 9, 2026Updated 8 months ago
- Humming is a high-performance, lightweight, and highly flexible JIT (Just-In-Time) compiled GEMM kernel library specifically designed for…☆250Updated this week
- ☆11Mar 9, 2022Updated 4 years ago
- ☆11Oct 13, 2023Updated 2 years ago
- An experimental communicating attention kernel based on DeepEP.☆34Jul 29, 2025Updated last year
- Github mirror of trition-lang/triton repo.☆201Updated this week
- ☆49Oct 23, 2025Updated 11 months ago
- [IPDPS 2024] Adaptive neighbor sampling for temporal GNN☆16Feb 17, 2025Updated last year
- Winner 🏆 (Agent-only) MLSys 2026 - FlashInfer AI Kernel Generation Contest for the DeepSeek Sparse Attention (DSA) track with an average…☆313Sep 12, 2026Updated 3 weeks ago
- Deploy on Railway without the complexity - Free Credits Offer • AdConnect your repo and Railway handles the rest with instant previews. Quickly provision container image services, databases, and storage volumes.
- Minimize server usage by leveraging a decentralized peer-to-peer network for ultra-low-latency live streaming among users.☆13Feb 19, 2024Updated 2 years ago
- 用pytorch训练18层残差神经网络，用pyqt设计界面☆12Jun 23, 2020Updated 6 years ago
- ☆12Mar 16, 2022Updated 4 years ago
- Cross-GPU KV Cache Marketplace☆27Nov 12, 2025Updated 10 months ago
- ☆19Mar 4, 2025Updated last year
- An unbounded n-gram language model on Tiny Shakespeare☆22Jan 21, 2026Updated 8 months ago
- ☆11Apr 20, 2020Updated 6 years ago
- A Scheduler for Batched LLM Inference☆19Oct 5, 2025Updated last year
- PyTorch Code for the Paper: "Exploiting Uncertainty of Loss Landscape for Stochastic Optimization [Bhaskara et al. (2019)]☆16Apr 30, 2026Updated 5 months ago
- Managed Kubernetes at scale on DigitalOcean • AdDigitalOcean Kubernetes includes the control plane, bandwidth allowance, container registry, automatic updates, and more for free.
- A pure-Python implementation of the Nvidia CuTe layout algebra intended to be approachable and easy to learn.☆253Jun 29, 2026Updated 3 months ago
- A Python DSL to write Nvidia PTX for Hopper and Blackwell in JAX and PyTorch☆384Jul 9, 2026Updated 3 months ago
- A Quirky Assortment of CuTe Kernels☆1,166Sep 17, 2026Updated 3 weeks ago
- IMMA-based **FP8-as-storage** GEMM experiments for Ampere (sm_86 / RTX 3090 Ti).☆28Jan 30, 2026Updated 8 months ago
- Next Generation Experimental Tracking for Machine Learning Operations☆18Sep 23, 2026Updated 2 weeks ago
- Composable and Embeddable Communication Runtime for Distributed AI Services☆101Updated this week
- 存放一些 CUDA 编程相关的博客文件。☆24Oct 16, 2025Updated 11 months ago