SkyworkAI / Skywork-MoELinks
Skywork-MoE: A Deep Dive into Training Techniques for Mixture-of-Experts Language Models
☆135Updated last year
Alternatives and similar repositories for Skywork-MoE
Users that are interested in Skywork-MoE are comparing it to the libraries listed below
Sorting:
- [ICML'24] The official implementation of “Rethinking Optimization and Architecture for Tiny Language Models”☆121Updated 5 months ago
- Mixture-of-Experts (MoE) Language Model☆189Updated 10 months ago
- Towards Economical Inference: Enabling DeepSeek's Multi-Head Latent Attention in Any Transformer-based LLMs☆178Updated 3 weeks ago
- ☆64Updated 7 months ago
- [ICML 2025] |TokenSwift: Lossless Acceleration of Ultra Long Sequence Generation☆110Updated last month
- A MoE impl for PyTorch, [ATC'23] SmartMoE☆64Updated 2 years ago
- ☆59Updated 3 weeks ago
- Repository of LV-Eval Benchmark☆67Updated 10 months ago
- ☆64Updated last month
- Exploring the Limit of Outcome Reward for Learning Mathematical Reasoning☆188Updated 3 months ago
- FuseAI Project☆87Updated 5 months ago
- Reformatted Alignment☆113Updated 9 months ago
- code for Scaling Laws of RoPE-based Extrapolation☆73Updated last year
- ☆56Updated 3 weeks ago
- A lightweight reinforcement learning framework that integrates seamlessly into your codebase, empowering developers to focus on algorithm…☆33Updated last month
- [EMNLP 2024] LongAlign: A Recipe for Long Context Alignment of LLMs☆250Updated 6 months ago
- Ouroboros: Speculative Decoding with Large Model Enhanced Drafting (EMNLP 2024 main)☆107Updated 3 months ago
- ☆109Updated last month
- The RedStone repository includes code for preparing extensive datasets used in training large language models.☆136Updated last week
- [COLM 2025] An Open Math Pre-trainng Dataset with 370B Tokens.☆95Updated 3 months ago
- [ICML 2025] Programming Every Example: Lifting Pre-training Data Quality Like Experts at Scale☆251Updated this week
- Inferflow is an efficient and highly configurable inference engine for large language models (LLMs).☆243Updated last year
- A highly capable 2.4B lightweight LLM using only 1T pre-training data with all details.☆195Updated last week
- Code for the preprint "Cache Me If You Can: How Many KVs Do You Need for Effective Long-Context LMs?"☆24Updated 3 weeks ago
- Counting-Stars (★)☆83Updated last month
- ☆107Updated last year
- Fast LLM Training CodeBase With dynamic strategy choosing [Deepspeed+Megatron+FlashAttention+CudaFusionKernel+Compiler];☆39Updated last year
- Implementation of the LongRoPE: Extending LLM Context Window Beyond 2 Million Tokens Paper☆146Updated 11 months ago
- Parameter-Efficient Sparsity Crafting From Dense to Mixture-of-Experts for Instruction Tuning on General Tasks☆144Updated 9 months ago
- This is a personal reimplementation of Google's Infini-transformer, utilizing a small 2b model. The project includes both model and train…☆58Updated last year