Qwen-7B

Training compute

10²³ FLOP

Parameters

7B

Published

Sep 28, 2023

Qwen-7B is an AI model developed by Alibaba (China), first published in September 2023. It works in the language domain, on tasks such as language modeling/generation and translation.

Training it took an estimated 10²³ FLOP of compute (estimation method: operation counting). The model has 7,000,000,000 parameters. It was trained on roughly 2.4T datapoints.

Access: Open weights (restricted use). Its weights are openly available. The reference paper has 3,775 citations. Epoch AI rates the confidence of this record as confident.

Full record

Organization: Alibaba
Country of organization: China
Domain: Language
Task: Language modeling/generation, Translation
Training compute: 10²³ FLOP
Compute estimation method: Operation counting
Parameters: 7,000,000,000
Dataset size: 2.4T
Model accessibility: Open weights (restricted use)
Open weights: Yes
Citations: 3,775
Epoch confidence: Confident

Benchmark results

01	Epoch Capabilities Index	ECI Score	106.5
02	PIQA	Score	0.78
03	BoolQ	Score	0.76
04	ARC AI2	Challenge score	0.75
05	LAMBADA	Score	0.68
06	GSM8K	EM	0.52
07	BIG-Bench Hard	Average	0.45
08	MMLU	EM	0.45