JT-Ushio / MHA2MLALinks
Towards Economical Inference: Enabling DeepSeek's Multi-Head Latent Attention in Any Transformer-based LLMs
☆178Updated 3 weeks ago
Alternatives and similar repositories for MHA2MLA
Users that are interested in MHA2MLA are comparing it to the libraries listed below
Sorting:
- Parallel Scaling Law for Language Model — Beyond Parameter and Inference Time Scaling☆410Updated last month
- TransMLA: Multi-Head Latent Attention Is All You Need☆327Updated last week
- Efficient triton implementation of Native Sparse Attention.☆175Updated last month
- [ICML 2025] |TokenSwift: Lossless Acceleration of Ultra Long Sequence Generation☆110Updated last month
- [ICLR 2025] DuoAttention: Efficient Long-Context LLM Inference with Retrieval and Streaming Heads☆473Updated 5 months ago
- Official implementation of "Fast-dLLM: Training-free Acceleration of Diffusion LLM by Enabling KV Cache and Parallel Decoding"☆278Updated last week
- Skywork-MoE: A Deep Dive into Training Techniques for Mixture-of-Experts Language Models☆135Updated last year
- ☆193Updated 2 months ago
- Chain of Experts (CoE) enables communication between experts within Mixture-of-Experts (MoE) models☆216Updated 2 weeks ago
- ☆109Updated last month
- Simple extension on vLLM to help you speed up reasoning model without training.☆166Updated last month
- Unofficial implementation for the paper "Mixture-of-Depths: Dynamically allocating compute in transformer-based language models"☆165Updated last year
- Rethinking RL Scaling for Vision Language Models: A Transparent, From-Scratch Framework and Comprehensive Evaluation Scheme☆133Updated 3 months ago
- Efficient LLM Inference over Long Sequences☆382Updated 2 weeks ago
- [ICLR 2025] COAT: Compressing Optimizer States and Activation for Memory-Efficient FP8 Training☆215Updated 3 weeks ago
- SeerAttention: Learning Intrinsic Sparse Attention in Your LLMs☆132Updated this week
- A highly capable 2.4B lightweight LLM using only 1T pre-training data with all details.☆195Updated last week
- Exploring the Limit of Outcome Reward for Learning Mathematical Reasoning☆188Updated 3 months ago
- 🔥 A minimal training framework for scaling FLA models☆188Updated last month
- qwen-nsa☆68Updated 3 months ago
- Official codebase for "Can 1B LLM Surpass 405B LLM? Rethinking Compute-Optimal Test-Time Scaling".☆268Updated 4 months ago
- Tina: Tiny Reasoning Models via LoRA☆266Updated last month
- OpenSeek aims to unite the global open source community to drive collaborative innovation in algorithms, data and systems to develop next…☆210Updated 2 weeks ago
- The official implementation of the paper <MoA: Mixture of Sparse Attention for Automatic Large Language Model Compression>☆139Updated last month
- Block Transformer: Global-to-Local Language Modeling for Fast Inference (NeurIPS 2024)☆158Updated 2 months ago
- [ICML'24] Data and code for our paper "Training-Free Long-Context Scaling of Large Language Models"☆421Updated 8 months ago
- The official repo of One RL to See Them All: Visual Triple Unified Reinforcement Learning☆287Updated last month
- [ICLR2025] Codebase for "ReMoE: Fully Differentiable Mixture-of-Experts with ReLU Routing", built on Megatron-LM.☆83Updated 6 months ago
- LongRoPE is a novel method that can extends the context window of pre-trained LLMs to an impressive 2048k tokens.☆232Updated 10 months ago
- VeOmni: Scaling any Modality Model Training to any Accelerators with PyTorch native Training Framework☆370Updated this week