ISEEKYAN/mbridge
Bridge Megatron-Core to Hugging Face/Reinforcement Learning
Repositories
ISEEKYAN/Megatron-LM
Ongoing research training transformer models at scale
ISEEKYAN/vllm
A high-throughput and memory-efficient inference and serving engine for LLMs
ISEEKYAN/verl-recipe
A set of examples based on verl for end-to-end RL training recipes.
ISEEKYAN/verl
verl: Volcano Engine Reinforcement Learning for LLMs
ISEEKYAN/FlashMLA
FlashMLA: Efficient Multi-head Latent Attention Kernels
ISEEKYAN/DeepGEMM
DeepGEMM: clean and efficient FP8 GEMM kernels with fine-grained scaling
ISEEKYAN/DeepEP
DeepEP: an efficient expert-parallel communication library
ISEEKYAN/verl_megatron_practice
(best/better) practices of megatron on veRL and tuning guide
ISEEKYAN/Megatron-Bridge
Training library for Megatron-based models
ISEEKYAN/k3
Megatron Lite support for moonshotai/Kimi-K3 (KDA + gated MLA hybrid attention, LatentMoE, MXFP4 weights) — external model integration example
ISEEKYAN/hy3
Standalone Tencent Hy3 support for Megatron-Lite — reference example of external model integration via register_model()
ISEEKYAN/miles
Miles is an enterprise-facing reinforcement learning framework for LLM and VLM post-training, forked from and co-evolving with slime.
ISEEKYAN/vime
An LLM post-training framework with vLLM for RL Scaling
ISEEKYAN/slime
slime is an LLM post-training framework for RL Scaling.
ISEEKYAN/ISEEKYAN.github.io
ISEEKYAN/Megatron-MoE-ModelZoo
Best practices for testing advanced Mixtral, DeepSeek, and Qwen series MoE models using Megatron Core MoE.
ISEEKYAN/fast-DiT
Fast Diffusion Models with Transformers