arcee-ai / Trinity-Large-Preview
Text Generation399B1.4K181Updated May 28, 2026
Downloads
1.4K
Likes
181
Last updated
May 28, 2026
License
openmdw-1.1
Trinity-Large-Preview is a 398B-parameter sparse Mixture-of-Experts (MoE) model with approximately 13B active parameters per token. It is the largest model in Arcee AI's Trinity family, trained on more than 17 trillion tokens and delivering frontier-level performance with stro…
Tasks & library
- text generation
- transformers
Tags
- transformers
- afmoe
- text-generation
- conversational
- custom_code
- en
Technical specifications
Parameters399B
Librarytransformers
Languages11 languages
Base modelarcee-ai/Trinity-Large-Base
Licenseopenmdw-1.1