Hunyuan

Training compute

1.2×10²⁴ FLOP

Parameters

100B

Published

Sep 7, 2023

Hunyuan is an AI model developed by Tencent (China), first published in September 2023. It works in the language, image generation and multimodal domain, on tasks such as language modeling/generation, image generation and question answering.

Training it took an estimated 1.2×10²⁴ FLOP of compute (estimation method: operation counting). The model has 100,000,000,000 parameters. It was trained on roughly 2T datapoints.

Access: API access. Its weights are not openly released. Epoch AI rates the confidence of this record as confident.

Full record

Organization: Tencent
Country of organization: China
Domain: Language, Image generation, Multimodal
Task: Language modeling/generation, Image generation, Question answering
Training compute: 1.2×10²⁴ FLOP
Compute estimation method: Operation counting
Parameters: 100,000,000,000
Dataset size: 2T
Model accessibility: API access
Open weights: No
Epoch confidence: Confident