Goblin
News
AI news by
promptgoblins.ai
|
News
About
News
About
Filtered by:
MoE
Clear
Titles
Summaries
8
Qwen Releases Open-Weight 125B MoE Model at One-Ninth the Training Cost of Its Predecessor
Models
2
5d ago
8
Qwen Releases Open-Weight 125B MoE Model at One-Ninth the Training Cost of Its Predecessor
Top
Models
· 2 srcs · 5d ago
Discuss
6
ToMoE: Technique Converts Dense LLMs to Mixture-of-Experts via Dynamic Pruning Without Permanent Parameter Loss
Research
1
Aug 24
6
ToMoE: Technique Converts Dense LLMs to Mixture-of-Experts via Dynamic Pruning Without Permanent Parameter Loss
Research
· 1 src · Aug 24
Discuss
6
Cursor Open-Sources Mixture-of-Kittens (MoK): MoE Training Megakernel Delivers 2.37x Speed Boost on NVIDIA Blackwell
Updated Aug 5
Research
3
Aug 5
6
Cursor Open-Sources Mixture-of-Kittens (MoK): MoE Training Megakernel Delivers 2.37x Speed Boost on NVIDIA Blackwell
Upd Aug 5
Research
· 3 srcs · Aug 5
Discuss
7
Swiftlet Runs 80B Qwen Model in 4.3 GB RAM on Mac and 35B on iPhone via MoE Expert Streaming
Research
1
Aug 4
7
Swiftlet Runs 80B Qwen Model in 4.3 GB RAM on Mac and 35B on iPhone via MoE Expert Streaming
Research
· 1 src · Aug 4
Discuss
7
RadixArk Open-Sources Miles: Scalable PyTorch-Native Framework for LLM Reinforcement Learning Post-Training
Open Source
1
Jul 1
7
RadixArk Open-Sources Miles: Scalable PyTorch-Native Framework for LLM Reinforcement Learning Post-Training
Open Source
· 1 src · Jul 1
Discuss
Filters
Signal
Title
Category
Sources
Posted
Discuss
