Open Issues Need Help
View All on GitHub [FEATURE] Fill the gaps in the embodied eval matrix about 1 month ago
documentation enhancement good first issue model
baidu-baige/LoongForge baidu-baige/LoongForge A modular, scalable, high-performance training framework for LLMs, VLMs, diffusion, and embodied models.
Open project →
313
A modular, scalable, high-performance training framework for LLMs, VLMs, diffusion, and embodied models.
Python
#ai#diffusion#distributed#dit#infra#llm#lora#megatron#mid-training#pretraining#sft#training#vla#vlm#wan
[FEATURE] Hope to support image generation models 4 months ago
enhancement good first issue model
baidu-baige/LoongForge baidu-baige/LoongForge A modular, scalable, high-performance training framework for LLMs, VLMs, diffusion, and embodied models.
Open project →
313
A modular, scalable, high-performance training framework for LLMs, VLMs, diffusion, and embodied models.
Python
#ai#diffusion#distributed#dit#infra#llm#lora#megatron#mid-training#pretraining#sft#training#vla#vlm#wan
enhancement good first issue
baidu-baige/LoongForge baidu-baige/LoongForge A modular, scalable, high-performance training framework for LLMs, VLMs, diffusion, and embodied models.
Open project →
313
A modular, scalable, high-performance training framework for LLMs, VLMs, diffusion, and embodied models.
Python
#ai#diffusion#distributed#dit#infra#llm#lora#megatron#mid-training#pretraining#sft#training#vla#vlm#wan
[FEATURE] ECHO: MoE Expert Load Balancing via Online Token Redistribution with Topology-Aware Dispatch Planning 5 months ago
enhancement good first issue
baidu-baige/LoongForge baidu-baige/LoongForge A modular, scalable, high-performance training framework for LLMs, VLMs, diffusion, and embodied models.
Open project →
313
A modular, scalable, high-performance training framework for LLMs, VLMs, diffusion, and embodied models.
Python
#ai#diffusion#distributed#dit#infra#llm#lora#megatron#mid-training#pretraining#sft#training#vla#vlm#wan