Alternatives and similar repositories for MHA2MLA

Users that are interested in MHA2MLA are comparing it to the libraries listed below

• QwenLM / ParScale
  Parallel Scaling Law for Language Model — Beyond Parameter and Inference Time Scaling
  ☆465Updated 7 months ago
• MuLabPKU / TransMLA
  TransMLA: Multi-Head Latent Attention Is All You Need (NeurIPS 2025 Spotlight)
  ☆422Updated 3 months ago
• step-law / steplaw
  ☆208Updated 2 months ago
• SkyworkAI / Skywork-MoE
  Skywork-MoE: A Deep Dive into Training Techniques for Mixture-of-Experts Language Models
  ☆138Updated last year
• mit-han-lab / duo-attention
  [ICLR 2025] DuoAttention: Efficient Long-Context LLM Inference with Retrieval and Streaming Heads
  ☆515Updated 10 months ago
• ZihanWang314 / CoE
  Chain of Experts (CoE) enables communication between experts within Mixture-of-Experts (MoE) models
  ☆227Updated last month
• thu-ml / ReMoE
  [ICLR2025] Codebase for "ReMoE: Fully Differentiable Mixture-of-Experts with ReLU Routing", built on Megatron-LM.
  ☆104Updated last year
• yaof20 / Flash-RL
  Implementation for FP8/INT8 Rollout for RL training without performence drop.
  ☆282Updated last month
• XunhaoLai / native-sparse-attention-triton
  Efficient triton implementation of Native Sparse Attention.
  ☆257Updated 7 months ago
• bigai-nlco / TokenSwift
  [ICML 2025] |TokenSwift: Lossless Acceleration of Ultra Long Sequence Generation
  ☆118Updated 7 months ago
• shangshang-wang / Tina
  Tina: Tiny Reasoning Models via LoRA
  ☆310Updated 3 months ago
• microsoft / SeerAttention
  SeerAttention: Learning Intrinsic Sparse Attention in Your LLMs
  ☆181Updated 3 months ago
• RUC-GSAI / YuLan-Mini
  A highly capable 2.4B lightweight LLM using only 1T pre-training data with all details.
  ☆222Updated 5 months ago
• hao-ai-lab / Dynasor
  [NeurIPS 2025] Simple extension on vLLM to help you speed up reasoning model without training.
  ☆215Updated 7 months ago
• NVlabs / COAT
  [ICLR 2025] COAT: Compressing Optimizer States and Activation for Memory-Efficient FP8 Training
  ☆256Updated 4 months ago
• OpenSparseLLMs / Linear-MoE
  ☆126Updated 6 months ago
• mdy666 / Qwen-Native-Sparse-Attention
  qwen-nsa
  ☆86Updated 2 months ago
• NVIDIA / Star-Attention
  Efficient LLM Inference over Long Sequences
  ☆394Updated 6 months ago
• stepfun-ai / Step3
  ☆440Updated 4 months ago
• astramind-ai / Mixture-of-depths
  Unofficial implementation for the paper "Mixture-of-Depths: Dynamically allocating compute in transformer-based language models"
  ☆176Updated last year
• OpenNLPLab / lightning-attention
  Lightning Attention-2: A Free Lunch for Handling Unlimited Sequence Lengths in Large Language Models
  ☆338Updated 10 months ago
• Tencent / AngelSlim
  Model compression toolkit engineered for enhanced usability, comprehensiveness, and efficiency.
  ☆236Updated this week
• stepfun-ai / PaCoRe
  PaCoRe: Learning to Scale Test-Time Compute with Parallel Coordinated Reasoning
  ☆222Updated 3 weeks ago
• RyanLiu112 / compute-optimal-tts
  Official codebase for "Can 1B LLM Surpass 405B LLM? Rethinking Compute-Optimal Test-Time Scaling".
  ☆278Updated 10 months ago
• jshuadvd / LongRoPE
  Implementation of the LongRoPE: Extending LLM Context Window Beyond 2 Million Tokens Paper
  ☆152Updated last year
• NVlabs / QeRL
  QeRL enables RL for 32B LLMs on a single H100 GPU.
  ☆470Updated last month
• xichen-fy / Fira
  Fira: Can We Achieve Full-rank Training of LLMs Under Low-rank Constraint?
  ☆119Updated last year
• weigao266 / Awesome-Efficient-Arch
  Speed Always Wins: A Survey on Efficient Architectures for Large Language Models
  ☆380Updated last month
• MiroMindAI / MiroMind-M1
  MiroMind-M1 is a fully open-source series of reasoning language models built on Qwen-2.5, focused on advancing mathematical reasoning.
  ☆246Updated 4 months ago
• Infini-AI-Lab / Multiverse
  ☆108Updated 3 months ago