dhcode-cpp / easy-dualpipeLinks
Pipeline-Parallel Lecture: Simplest Dualpipe Implementation.
☆27Updated 2 months ago
Alternatives and similar repositories for easy-dualpipe
Users that are interested in easy-dualpipe are comparing it to the libraries listed below
Sorting:
- ☆146Updated 5 months ago
- Tiny-DeepSpeed, a minimalistic re-implementation of the DeepSpeed library☆42Updated last week
- 青稞Talk☆135Updated this week
- ☆143Updated last month
- Efficient, Flexible, and Highly Fault-Tolerant Model Service Management Based on SGLang☆56Updated 9 months ago
- simplest online-softmax notebook for explain Flash Attention☆13Updated 8 months ago
- 注释的nano_vllm仓库,并且完成了MiniCPM4的适配以及注册新模型的功能☆39Updated 2 weeks ago
- UltraScale Playbook 中文版☆71Updated 5 months ago
- ☆113Updated 9 months ago
- A flexible and efficient training framework for large-scale alignment tasks☆415Updated this week
- Implement some method of LLM KV Cache Sparsity☆35Updated last year
- Inference code for LLaMA models☆122Updated 2 years ago
- ☆79Updated last year
- A MoE impl for PyTorch, [ATC'23] SmartMoE☆67Updated 2 years ago
- DashInfer is a native LLM inference engine aiming to deliver industry-leading performance atop various hardware architectures, including …☆263Updated 3 weeks ago
- Model compression toolkit engineered for enhanced usability, comprehensiveness, and efficiency.☆89Updated this week
- Tiny-Megatron, a minimalistic re-implementation of the Megatron library☆16Updated last week
- DeepSeek Native Sparse Attention pytorch implementation☆91Updated 3 weeks ago
- Triton Documentation in Chinese Simplified / Triton 中文文档☆80Updated 4 months ago
- [ICLR 2025] PEARL: Parallel Speculative Decoding with Adaptive Draft Length☆108Updated 4 months ago
- FlagScale is a large model toolkit based on open-sourced projects.☆347Updated this week
- Implementation of FlashAttention in PyTorch☆162Updated 7 months ago
- Compare different hardware platforms via the Roofline Model for LLM inference tasks.☆112Updated last year
- ☆54Updated this week
- ☆52Updated 2 years ago
- ☆42Updated last year
- ☆78Updated 4 months ago
- CPM.cu is a lightweight, high-performance CUDA implementation for LLMs, optimized for end-device inference and featuring cutting-edge tec…☆175Updated this week
- 从零构建MoE大模型:从预训练到DPO的完整实践☆52Updated this week
- qwen-nsa☆74Updated 4 months ago