Hunyuan-Large

Training compute

3.5×10²⁴ FLOP

Parameters

389B

Published

Nov 6, 2024

Hunyuan-Large is an AI model developed by Tencent (China), first published in November 2024. It works in the language domain, on tasks such as language modeling/generation, question answering, code generation and translation.

Training it took an estimated 3.5×10²⁴ FLOP of compute (estimation method: operation counting). The model has 389,000,000,000 parameters. It was trained on roughly 7T datapoints.

Access: Open weights (restricted use). Its weights are openly available. Epoch AI rates the confidence of this record as confident.

Full record

Organization: Tencent
Country of organization: China
Domain: Language
Task: Language modeling/generation, Question answering, Code generation, Translation
Training compute: 3.5×10²⁴ FLOP
Compute estimation method: Operation counting
Parameters: 389,000,000,000
Dataset size: 7T
Numerical format: BF16
Model accessibility: Open weights (restricted use)
Open weights: Yes
Epoch confidence: Confident