Qwen3-0.6B megakernel: 527 tok/s decode on RTX 3090 (3.8x faster than PyTorch)
☆98Feb 10, 2026Updated 2 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:
- EleutherAI ML Performance reading group repository (slides, meeting recordings, annotated papers)☆31Mar 20, 2026Updated last month
- ☆25Apr 23, 2026Updated last week
- ☆14Aug 28, 2019Updated 6 years ago
- Code for "Accelerating Training with Neuron Interaction and Nowcasting Networks" [ICLR 2025]☆29Feb 20, 2026Updated 2 months ago
- Benchmark tests supporting the TiledCUDA library.☆18Nov 19, 2024Updated last year
- Deploy to Railway using AI coding agents - Free Credits Offer • AdUse Claude Code, Codex, OpenCode, and more. Autonomous software development now has the infrastructure to match with Railway.
- Backend for the diffusion-ui frontend☆25Feb 17, 2024Updated 2 years ago
- Jax implementation of the AdaHessian optimizer☆20Mar 11, 2021Updated 5 years ago
- Created a simple neural network using C++17 standard and the Eigen library that supports both forward and backward propagation.☆11Jul 27, 2024Updated last year
- Simple and efficient memory pool is implemented with C++11.☆10Jun 2, 2022Updated 3 years ago
- Manifold-Mixup implementation for fastai V1☆19Oct 1, 2020Updated 5 years ago
- 数据库内核笔记☆13Aug 18, 2022Updated 3 years ago
- A collection of optimizers, some arcane others well known, for Flax.☆29Aug 6, 2021Updated 4 years ago
- GEMV implementation with CUTLASS☆21Aug 21, 2025Updated 8 months ago
- ☆17Dec 11, 2022Updated 3 years ago
- Deploy to Railway using AI coding agents - Free Credits Offer • AdUse Claude Code, Codex, OpenCode, and more. Autonomous software development now has the infrastructure to match with Railway.
- ☆11May 2, 2023Updated 3 years ago
- Flash Attention in ~100 lines of CUDA (forward pass only)☆12Jun 10, 2024Updated last year
- NanoGPT-speedrunning for the poor T4 enjoyers☆73Apr 22, 2025Updated last year
- High Performance FP8 GEMM Kernels for SM89 and later GPUs.☆21Jan 24, 2025Updated last year
- -☆11Nov 21, 2020Updated 5 years ago
- Row-wise block scaling for fp8 quantization matrix multiplication. Solution to GPU mode AMD challenge.☆19Feb 9, 2026Updated 2 months ago
- TensorRT-in-Action 是一个 GitHub 代码库，提供了使用 TensorRT 的代码示例，并有对应 Jupyter Notebook。☆15Jun 1, 2023Updated 2 years ago
- Code for AAAI 2023 accepted paper titled "Knowledge-Bridged Causal Interaction Network for Causal Emotion Entailment"☆14May 6, 2023Updated 2 years ago
- Pytorch implementation of "Oscillation-Reduced MXFP4 Training for Vision Transformers" on DeiT Model Pre-training☆39Jun 20, 2025Updated 10 months ago
- Wordpress hosting with auto-scaling - Free Trial Offer • AdFully Managed hosting for WordPress and WooCommerce businesses that need reliable, auto-scalable performance. Cloudways SafeUpdates now available.
- Tensara's GPU programming problems☆19Apr 23, 2026Updated last week
- Codes and data for ACL 2023 Findings paper "Click: Controllable Text Generation with Sequence Likelihood Contrastive Learning"☆19Feb 26, 2024Updated 2 years ago
- [EMNLP 2023] Generative Emotion Cause Triplet Extraction in Conversations with Commonsense Knowledge☆14Jan 5, 2024Updated 2 years ago
- 🚥 Quickly set the style of Anybar☆14Jun 17, 2021Updated 4 years ago
- 方便扩展的Cuda算子理解和优化框架，仅用在学习使用☆18Jun 13, 2024Updated last year
- This is a repo covers ai research papers pseudocodes☆17Jun 20, 2023Updated 2 years ago
- ☆23Aug 14, 2024Updated last year
- ☆83Mar 3, 2026Updated 2 months ago
- An example platform daemon in Rust; written for Mastering Embedded Linux☆12May 8, 2020Updated 5 years ago
- Deploy to Railway using AI coding agents - Free Credits Offer • AdUse Claude Code, Codex, OpenCode, and more. Autonomous software development now has the infrastructure to match with Railway.
- 《C++模板元编程实战：一个深度学习框架的初步实现》记录。☆18Nov 5, 2022Updated 3 years ago
- High-performance KV cache storage for LLM inference — GPU offloading, SSD caching, and cross-node sharing via RDMA. Works with vLLM and S…☆51Apr 28, 2026Updated last week
- PolEval 2021 Task 1☆15Jun 28, 2022Updated 3 years ago
- GPU-Accelerated Cosine Similarity for Tandem Mass Spectrometry☆17Nov 4, 2025Updated 6 months ago
- Costum menu bar entry for MacOS - build with rumps and py2app.☆13Oct 4, 2023Updated 2 years ago
- GEMM by WMMA (tensor core)☆15Jul 31, 2022Updated 3 years ago
- A combination of Oobabooga's fork and the main cuda branch of GPTQ-for-LLaMa in a package format.☆23Oct 6, 2023Updated 2 years ago