Towards Economical Inference: Enabling DeepSeek's Multi-Head Latent Attention in Any Transformer-based LLMs
☆212Dec 4, 2025Updated 8 months ago
Alternatives and similar repositories for MHA2MLA
Users that are interested in MHA2MLA are comparing it to the libraries listed below. We may earn a commission when you buy through links labeled 'Ad' on this page.
Sorting:
- Design hardware-friendly model architectures and migrate existing LLMs with minimal performance loss☆499Jul 14, 2026Updated last month
- CutDiffusion: A Simple, Fast, Cheap, and Strong Diffusion Extrapolation Method☆27Oct 9, 2025Updated 10 months ago
- Block Transformer: Global-to-Local Language Modeling for Fast Inference (NeurIPS 2024)☆167Apr 13, 2025Updated last year
- an open source reproduction of NVIDIA's nGPT (Normalized Transformer with Representation Learning on the Hypersphere)☆112Mar 7, 2025Updated last year
- Efficient triton implementation of Native Sparse Attention.☆283May 23, 2025Updated last year
- GPU virtual machines on DigitalOcean Gradient AI • AdGet to production fast with high-performance AMD and NVIDIA GPUs you can spin up in seconds. The definition of operational simplicity.
- code for Scaling Laws of RoPE-based Extrapolation☆73Oct 16, 2023Updated 2 years ago
- 🐳 Efficient Triton implementations for "Native Sparse Attention: Hardware-Aligned and Natively Trainable Sparse Attention"☆1,018Feb 5, 2026Updated 6 months ago
- Repository for "TESS-2: A Large-Scale, Generalist Diffusion Language Model"☆58Feb 20, 2025Updated last year
- Layer-Condensed KV cache w/ 10 times larger batch size, fewer params and less computation. Dramatic speed up with better task performance…☆156Apr 7, 2025Updated last year
- ☆47Aug 29, 2024Updated last year
- Fast and Slow Generating: An Empirical Study on Large and Small Language Models Collaborative Decoding.☆13Nov 19, 2024Updated last year
- Skywork-MoE: A Deep Dive into Training Techniques for Mixture-of-Experts Language Models☆140Jun 12, 2024Updated 2 years ago
- MoBA: Mixture of Block Attention for Long-Context LLMs☆2,166Apr 3, 2025Updated last year
- The official implementation of paper: SimLayerKV: A Simple Framework for Layer-Level KV Cache Reduction.☆55Oct 18, 2024Updated last year
- Proton VPN Special Offer - Get 70% off • AdSpecial partner offer. Trusted by over 100 million users worldwide. Tested, Approved and Recommended by Experts.
- Code for the paper "Function-Space Learning Rates"☆23Jun 3, 2025Updated last year
- An efficient implementation of the NSA (Native Sparse Attention) kernel☆135Jun 24, 2025Updated last year
- [NeurIPS'24 Spotlight, ICLR'25, ICML'25] To speed up Long-context LLMs' inference, approximate and dynamic sparse calculate the attention…☆1,226Apr 8, 2026Updated 4 months ago
- [NeurIPS 2024] Official Repository of The Mamba in the Llama: Distilling and Accelerating Hybrid Models☆243Oct 14, 2025Updated 10 months ago
- ☆127Jun 2, 2026Updated 2 months ago
- The official code repo and data hub of top_nsigma sampling strategy for LLMs.☆27Feb 11, 2025Updated last year
- Muon is Scalable for LLM Training☆1,534Aug 3, 2025Updated last year
- [ICML2025] SpargeAttention: A training-free sparse attention that accelerates any model inference.☆1,030Feb 25, 2026Updated 5 months ago
- FlashMLA: Efficient Multi-head Latent Attention Kernels☆12,853Jul 28, 2026Updated 3 weeks ago
- Deploy to Railway using AI coding agents - Free Credits Offer • AdUse Claude Code, Codex, OpenCode, and more. Autonomous software development now has the infrastructure to match with Railway.
- Code for paper: [ICLR2025 Oral] FlexPrefill: A Context-Aware Sparse Attention Mechanism for Efficient Long-Sequence Inference☆170Oct 13, 2025Updated 10 months ago
- GEAR: An Efficient KV Cache Compression Recipefor Near-Lossless Generative Inference of LLM☆186Jul 12, 2024Updated 2 years ago
- ☆138May 29, 2025Updated last year
- 🚀 Efficient implementations for emerging model architectures☆5,575Updated this week
- An unofficial implementation of "Mixture-of-Depths: Dynamically allocating compute in transformer-based language models"☆35Jun 7, 2024Updated 2 years ago
- [ICLR2025] Breaking Throughput-Latency Trade-off for Long Sequences with Speculative Decoding☆156Dec 4, 2024Updated last year
- ☆23Sep 19, 2024Updated last year
- Rectified Rotary Position Embeddings☆398May 20, 2024Updated 2 years ago
- [ASPLOS'26] Taming the Long-Tail: Efficient Reasoning RL Training with Adaptive Drafter☆176Feb 27, 2026Updated 5 months ago
- Deploy to Railway using AI coding agents - Free Credits Offer • AdUse Claude Code, Codex, OpenCode, and more. Autonomous software development now has the infrastructure to match with Railway.
- Fork of Flame repo for training of some new stuff in development☆20Aug 10, 2026Updated last week
- (ACL2025 oral) SCOPE: Optimizing KV Cache Compression in Long-context Generation☆36May 28, 2025Updated last year
- ☆171Feb 15, 2025Updated last year
- ☆156Mar 4, 2025Updated last year
- ☆192Jan 14, 2025Updated last year
- The source code of our work "Prepacking: A Simple Method for Fast Prefilling and Increased Throughput in Large Language Models" [AISTATS …☆62Oct 11, 2024Updated last year
- Lightning Attention-2: A Free Lunch for Handling Unlimited Sequence Lengths in Large Language Models☆344Feb 23, 2025Updated last year