LLM EXPLORER 60,277 MODELS INDEXED

TR HASH MoE 100M 125B Agentic Pretraining by AETHORIA-AI

By AETHORIA-AI · 603 downloads

TR HASH MoE 100M 125B Agentic Pretraining is an open-source language model by AETHORIA-AI. Features: 125b LLM, Context: 2K, License: cc-by-nc-4.0, MoE, LLM Explorer Score: 0.32.

  Agentic-pretraining   Custom-code Dataset:aethoria-ai/tr-hash-pr...   Deterministic-routing   En   Fr   Gqa   Mixture-of-experts   Moe   Multi-hash-routing   Pytorch   Region:us   Safetensors   Tr-hash   Tr hash moe

TR HASH MoE 100M 125B Agentic Pretraining Parameters and Internals

LLM NameTR HASH MoE 100M 125B Agentic Pretraining
Repository πŸ€—https://huggingface.co/AETHORIA-AI/TR-HASH-MoE-100M-125B-Agentic-Pretraining 
Model Size125b
Updated2026-09-03
MaintainerAETHORIA-AI
Model Typetr_hash_moe
Supported Languagesen fr
Model ArchitectureTRHashForCausalLM
Licensecc-by-nc-4.0
Context Length2048
Model Max Length2048
Vocabulary Size32000
Torch Data Typebfloat16