deepseek-ai / FlashMLALinks
FlashMLA: Efficient Multi-head Latent Attention Kernels
☆11,819Updated 3 weeks ago
Alternatives and similar repositories for FlashMLA
Users that are interested in FlashMLA are comparing it to the libraries listed below
Sorting:
- DeepEP: an efficient expert-parallel communication library☆8,630Updated this week
- DeepGEMM: clean and efficient FP8 GEMM kernels with fine-grained scaling☆5,812Updated last week
- A bidirectional pipeline parallelism algorithm for computation-communication overlap in DeepSeek V3/R1 training.☆2,869Updated 7 months ago
- Production-tested AI infrastructure tools for efficient AGI development and community-driven innovation☆7,923Updated 5 months ago
- Mooncake is the serving platform for Kimi, a leading LLM service provided by Moonshot AI.☆4,124Updated this week
- Expert Parallelism Load Balancer☆1,279Updated 6 months ago
- A high-performance distributed file system designed to address the challenges of AI training and inference workloads.☆9,386Updated last week
- Analyze computation-communication overlap in V3/R1.☆1,105Updated 7 months ago
- A Flexible Framework for Experiencing Cutting-edge LLM Inference Optimizations☆15,207Updated this week
- SGLang is a fast serving framework for large language models and vision language models.☆19,094Updated this week
- s1: Simple test-time scaling☆6,576Updated 3 months ago
- Lightning-Fast RL for LLM Reasoning and Agents. Made Simple & Flexible.☆2,853Updated this week
- A Datacenter Scale Distributed Inference Serving Framework☆5,325Updated this week
- The official repo of MiniMax-Text-01 and MiniMax-VL-01, large-language-model & vision-language-model based on Linear Attention☆3,197Updated 3 months ago
- FlashInfer: Kernel Library for LLM Serving☆3,952Updated this week
- verl: Volcano Engine Reinforcement Learning for LLMs☆14,648Updated this week
- Minimal reproduction of DeepSeek R1-Zero☆12,278Updated 5 months ago
- MoBA: Mixture of Block Attention for Long-Context LLMs☆1,941Updated 6 months ago
- Domain-specific language designed to streamline the development of high-performance GPU/CPU/Accelerators kernels☆3,658Updated this week
- Qwen3-VL is the multimodal large language model series developed by Qwen team, Alibaba Cloud.☆15,013Updated last week
- Democratizing Reinforcement Learning for LLMs☆4,534Updated this week
- Simple RL training for reasoning☆3,773Updated 2 months ago
- High-performance inference framework for large language models, focusing on efficiency, flexibility, and availability.☆1,307Updated this week
- Qwen2.5-Omni is an end-to-end multimodal model by Qwen team at Alibaba Cloud, capable of understanding text, audio, vision, video, and pe…☆3,728Updated 4 months ago
- An Easy-to-use, Scalable and High-performance RLHF Framework based on Ray (PPO & GRPO & REINFORCE++ & vLLM & Ray & Dynamic Sampling & Asy…☆8,180Updated 2 weeks ago
- Nano vLLM☆7,173Updated last month
- ☆1,583Updated 10 months ago
- Qwen3 is the large language model series developed by Qwen team, Alibaba Cloud.☆25,085Updated last week
- Reproduce R1 Zero on Logic Puzzle☆2,407Updated 7 months ago
- Community maintained hardware plugin for vLLM on Ascend☆1,230Updated this week