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
  ☆463Updated 7 months ago
• step-law / steplaw
  ☆208Updated last month
• SkyworkAI / Skywork-MoE
  Skywork-MoE: A Deep Dive into Training Techniques for Mixture-of-Experts Language Models
  ☆137Updated last year
• mit-han-lab / duo-attention
  [ICLR 2025] DuoAttention: Efficient Long-Context LLM Inference with Retrieval and Streaming Heads
  ☆515Updated 10 months ago
• MuLabPKU / TransMLA
  TransMLA: Multi-Head Latent Attention Is All You Need (NeurIPS 2025 Spotlight)
  ☆419Updated 3 months ago
• NVlabs / COAT
  [ICLR 2025] COAT: Compressing Optimizer States and Activation for Memory-Efficient FP8 Training
  ☆252Updated 4 months ago
• bigai-nlco / TokenSwift
  [ICML 2025] |TokenSwift: Lossless Acceleration of Ultra Long Sequence Generation
  ☆118Updated 7 months ago
• thu-ml / ReMoE
  [ICLR2025] Codebase for "ReMoE: Fully Differentiable Mixture-of-Experts with ReLU Routing", built on Megatron-LM.
  ☆102Updated last year
• OpenSparseLLMs / Linear-MoE
  ☆126Updated 6 months ago
• yaof20 / Flash-RL
  Implementation for FP8/INT8 Rollout for RL training without performence drop.
  ☆281Updated last month
• xichen-fy / Fira
  Fira: Can We Achieve Full-rank Training of LLMs Under Low-rank Constraint?
  ☆119Updated last year
• ZihanWang314 / CoE
  Chain of Experts (CoE) enables communication between experts within Mixture-of-Experts (MoE) models
  ☆226Updated last month
• astramind-ai / Mixture-of-depths
  Unofficial implementation for the paper "Mixture-of-Depths: Dynamically allocating compute in transformer-based language models"
  ☆176Updated last year
• Tencent / AngelSlim
  Model compression toolkit engineered for enhanced usability, comprehensiveness, and efficiency.
  ☆228Updated this week
• OpenNLPLab / lightning-attention
  Lightning Attention-2: A Free Lunch for Handling Unlimited Sequence Lengths in Large Language Models
  ☆338Updated 10 months ago
• RUC-GSAI / YuLan-Mini
  A highly capable 2.4B lightweight LLM using only 1T pre-training data with all details.
  ☆222Updated 4 months ago
• microsoft / SeerAttention
  SeerAttention: Learning Intrinsic Sparse Attention in Your LLMs
  ☆177Updated 3 months ago
• hao-ai-lab / Dynasor
  [NeurIPS 2025] Simple extension on vLLM to help you speed up reasoning model without training.
  ☆215Updated 6 months ago
• stepfun-ai / Step3
  ☆440Updated 4 months ago
• XunhaoLai / native-sparse-attention-triton
  Efficient triton implementation of Native Sparse Attention.
  ☆256Updated 7 months ago
• pprp / Awesome-Efficient-MoE
  Efficient Mixture of Experts for LLM Paper List
  ☆149Updated 2 months ago
• OpenBMB / infllmv2_cuda_impl
  ☆78Updated 3 weeks ago
• mdy666 / Qwen-Native-Sparse-Attention
  qwen-nsa
  ☆85Updated 2 months ago
• shangshang-wang / Tina
  Tina: Tiny Reasoning Models via LoRA
  ☆310Updated 3 months ago
• FasterDecoding / TEAL
  ☆156Updated 10 months ago
• NVIDIA / Star-Attention
  Efficient LLM Inference over Long Sequences
  ☆393Updated 5 months ago
• inclusionAI / Ling-V2
  Ling-V2 is a MoE LLM provided and open-sourced by InclusionAI.
  ☆248Updated 2 months ago
• SalesforceAIResearch / GemFilter
  ☆85Updated last month
• Dao-AILab / grouped-latent-attention
  ☆133Updated 6 months ago
• microsoft / LongRoPE
  LongRoPE is a novel method that can extends the context window of pre-trained LLMs to an impressive 2048k tokens.
  ☆276Updated last month