fzyzcjy / torch_utilsLinks
Utility scripts for PyTorch (e.g. Memory profiler that understands more low-level allocations such as NCCL)
☆16Updated 3 weeks ago
Alternatives and similar repositories for torch_utils
Users that are interested in torch_utils are comparing it to the libraries listed below
Sorting:
- ☆49Updated 2 months ago
- ☆60Updated 3 months ago
- DeeperGEMM: crazy optimized version☆71Updated 2 months ago
- Quantized Attention on GPU☆44Updated 8 months ago
- Framework to reduce autotune overhead to zero for well known deployments.☆79Updated last week
- Bridge Megatron-Core to Hugging Face/Reinforcement Learning☆67Updated this week
- Odysseus: Playground of LLM Sequence Parallelism☆72Updated last year
- A Suite for Parallel Inference of Diffusion Transformers (DiTs) on multi-GPU Clusters☆47Updated last year
- ☆75Updated 2 months ago
- An experimental communicating attention kernel based on DeepEP.☆32Updated this week
- Debug print operator for cudagraph debugging☆13Updated last year
- A simple calculation for LLM MFU.☆42Updated 5 months ago
- Accelerate LLM preference tuning via prefix sharing with a single line of code☆42Updated last month
- DeepXTrace is a lightweight tool for precisely diagnosing slow ranks in DeepEP-based environments.☆38Updated this week
- ☆78Updated 3 months ago
- PyTorch bindings for CUTLASS grouped GEMM.☆106Updated 2 months ago
- TileFusion is an experimental C++ macro kernel template library that elevates the abstraction level in CUDA C for tile processing.☆91Updated last month
- Estimate MFU for DeepSeekV3☆25Updated 6 months ago
- GPTQ inference TVM kernel☆40Updated last year
- ☆44Updated this week
- Automated Parallelization System and Infrastructure for Multiple Ecosystems☆79Updated 8 months ago
- ☆38Updated 3 months ago
- Best practices for testing advanced Mixtral, DeepSeek, and Qwen series MoE models using Megatron Core MoE.☆41Updated this week
- ☆85Updated 8 months ago
- Implement Flash Attention using Cute.☆92Updated 7 months ago
- TritonBench: Benchmarking Large Language Model Capabilities for Generating Triton Operators☆70Updated last month
- ⚡️Write HGEMM from scratch using Tensor Cores with WMMA, MMA and CuTe API, Achieve Peak⚡️ Performance.☆93Updated 2 months ago
- ☆92Updated 4 months ago
- Tritonbench is a collection of PyTorch custom operators with example inputs to measure their performance.☆193Updated this week
- ☆96Updated 10 months ago