harleyszhang / llm_countsLinks
llm theoretical performance analysis tools and support params, flops, memory and latency analysis.
☆96Updated last week
Alternatives and similar repositories for llm_counts
Users that are interested in llm_counts are comparing it to the libraries listed below
Sorting:
- 使用 CUDA C++ 实现的 llama 模型推理框架☆58Updated 7 months ago
- A light llama-like llm inference framework based on the triton kernel.☆128Updated this week
- Examples of CUDA implementations by Cutlass CuTe☆195Updated 4 months ago
- Summary of some awesome work for optimizing LLM inference☆76Updated 2 weeks ago
- Optimize softmax in triton in many cases☆21Updated 9 months ago
- Implement Flash Attention using Cute.☆87Updated 6 months ago
- ☆148Updated 5 months ago
- ☆135Updated last year
- 使用 cutlass 仓库在 ada 架构上实现 fp8 的 flash attention☆70Updated 10 months ago
- ☆77Updated last month
- ⚡️Write HGEMM from scratch using Tensor Cores with WMMA, MMA and CuTe API, Achieve Peak⚡️ Performance.☆79Updated last month
- A lightweight design for computation-communication overlap.☆141Updated last week
- 使用 cutlass 实现 flash-attention 精简版,具有教学意义☆41Updated 10 months ago
- ☆141Updated 3 months ago
- ☆27Updated last month
- [USENIX ATC '24] Accelerating the Training of Large Language Models using Efficient Activation Rematerialization and Optimal Hybrid Paral…☆55Updated 10 months ago
- High performance Transformer implementation in C++.☆125Updated 5 months ago
- learning how CUDA works☆269Updated 3 months ago
- A minimalist and extensible PyTorch extension for implementing custom backend operators in PyTorch.☆33Updated last year
- ☆58Updated 7 months ago
- ☆139Updated last year
- ☆96Updated 9 months ago
- Performance of the C++ interface of flash attention and flash attention v2 in large language model (LLM) inference scenarios.☆38Updated 3 months ago
- ☆122Updated 6 months ago
- A tutorial for CUDA&PyTorch☆144Updated 5 months ago
- ⚡️FFPA: Extend FlashAttention-2 with Split-D, achieve ~O(1) SRAM complexity for large headdim, 1.8x~3x↑ vs SDPA.☆186Updated last month
- ☆97Updated 2 months ago
- A standalone GEMM kernel for fp16 activation and quantized weight, extracted from FasterTransformer☆92Updated 3 weeks ago
- hands on model tuning with TVM and profile it on a Mac M1, x86 CPU, and GTX-1080 GPU.☆48Updated 2 years ago
- ☆36Updated 10 months ago