Qwen3-0.6B megakernel: 527 tok/s decode on RTX 3090 (3.8x faster than PyTorch)
☆145Feb 10, 2026Updated 8 months ago
Alternatives and similar repositories for MegaQwen
Users that are interested in MegaQwen are comparing it to the libraries listed below. We may earn a commission when you buy through links labeled 'Ad' on this page.
Sorting:
- Aggressive decode optimizations for Qwen3-0.6B on RTX 5090☆64Feb 25, 2026Updated 8 months ago
- Tile primitives for speedy kernels☆21Sep 13, 2026Updated last month
- LLM Inference via Triton (Flexible & Modular): Focused on Kernel Optimization using CUBIN binaries, Starting from gpt-oss Model☆119Apr 28, 2026Updated 6 months ago
- Kernels, of the mega variety :)☆875May 26, 2026Updated 5 months ago
- An agent harness that compiles a model into one provably-correct, self-retargeting CUDA megakernel and self-tunes it past cuBLAS at batch…☆157Sep 18, 2026Updated last month
- 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.
- High-performance GPU kernels for LLM inference in OpenAI Triton. Fused RMSNorm, SwiGLU, INT8 GEMM with benchmarks and roofline analysis.☆44Jul 22, 2026Updated 3 months ago
- Optimize GEMM with tensorcore step by step☆40Dec 17, 2023Updated 2 years ago
- Implemented and benchmarked KV cache compression methods for LLM inference in Triton. Optimized kernels for KIVI and TurboQuant with Flas…☆18Jun 19, 2026Updated 4 months ago
- Writing FLUX in Triton☆42Sep 22, 2024Updated 2 years ago
- mKernel: fast multi-node, multi-GPU fused kernels☆306Oct 17, 2026Updated last week
- Pure Triton kernels for Qwen3.5-27B inference on NVIDIA B200☆124Feb 28, 2026Updated 8 months ago
- ☆30Oct 15, 2026Updated 2 weeks ago
- A simple environment for writing and experimenting with hand-written CUDA PTX kernels.☆19Sep 11, 2025Updated last year
- ☆20Apr 26, 2026Updated 6 months ago
- AI Agents on DigitalOcean Gradient AI Platform • AdBuild production-ready AI agents using customizable tools or access multiple LLMs through a single endpoint. Create custom knowledge bases or connect external data.
- Speed-of-Light SW efficiency by using ultra low-latency primitives for comms collectives☆13Apr 8, 2026Updated 6 months ago
- ☆98Mar 26, 2025Updated last year
- ☆55Updated this week
- ☆40Updated this week
- Humming is a high-performance, lightweight, and highly flexible JIT (Just-In-Time) compiled GEMM kernel library specifically designed for…☆337Updated this week
- A PyTorch implementation of the GPT-OSS-20B architecture. All components are coded from scratch: RoPE with YaRN, RMSNorm, SwiGLU with cla…☆238Dec 2, 2025Updated 10 months ago
- Fastest kernels written from scratch☆629Aug 15, 2026Updated 2 months ago
- CODA: Rewriting Transformer Blocks as GEMM-Epilogue Programs☆258Updated this week
- High-performance LLM operator library built on TileLang.☆202Updated this week
- 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.
- ☆534Aug 26, 2026Updated 2 months ago
- A series of high-performance GEMM (General Matrix Multiply) implementations Iteratively optimised for H100 GPUs in Pure CUDA.☆83Feb 18, 2026Updated 8 months ago
- A sparse-first inference engine (SparseEngine).☆106Updated this week
- ☆37Aug 31, 2026Updated last month
- writing really fast kernels☆21Oct 11, 2026Updated 2 weeks ago
- MetaAttention: A Unified and Performant Attention Framework Across Hardware Backends(PPoPP'26)☆18Aug 6, 2026Updated 2 months ago
- TileFusion is an experimental C++ macro kernel template library that elevates the abstraction level in CUDA C for tile processing.☆119Aug 4, 2026Updated 2 months ago
- GGML bindings that aim to be idiomatic Rust rather than directly corresponding to the C/C++ interface☆20Sep 25, 2023Updated 3 years ago
- [NeurIPS 2025] ClusterFusion: Expanding Operator Fusion Scope for LLM Inference via Cluster-Level Collective Primitive☆79Aug 8, 2026Updated 2 months ago
- Managed hosting for WordPress and PHP on Cloudways • AdManaged hosting for WordPress, Magento, Laravel, or PHP apps, on multiple cloud providers. Deploy in minutes on Cloudways by DigitalOcean.
- Mirage Persistent Kernel: Compiling LLMs into a MegaKernel☆2,582Updated this week
- Optimize SGEMM step by step☆10Sep 4, 2024Updated 2 years ago
- 存放一些 CUDA 编程相关的博客文件。☆23Oct 16, 2025Updated last year
- high-performance linear attention kernel library built on TileLang☆742Oct 14, 2026Updated 2 weeks ago
- ☆15Jul 10, 2026Updated 3 months ago
- Github mirror of trition-lang/triton repo.☆205Updated this week
- Region-level profiling for CUDA kernels with trace, NVBit, CUPTI, NSys, and an interactive Explorer.☆133Apr 17, 2026Updated 6 months ago