Llama 2-13B

Training compute

1.6×10²³ FLOP

Parameters

13B

Published

Jul 18, 2023

Llama 2-13B is an AI model developed by Meta AI (United States), first published in July 2023. It works in the language domain, on tasks such as language modeling.

Training it took an estimated 1.6×10²³ FLOP of compute (estimation method: hardware,operation counting). The model has 13,000,000,000 parameters. It was trained on roughly 2T datapoints. Training ran on NVIDIA A100 SXM4 80 GB. The compute alone is estimated at $235K in 2023 dollars.

Access: Open weights (restricted use). Its weights are openly available. The reference paper has 16,911 citations. Epoch AI rates the confidence of this record as confident.

Full record

Organization: Meta AI
Country of organization: United States
Domain: Language
Task: Language modeling
Training compute: 1.6×10²³ FLOP
Compute estimation method: Hardware, Operation counting
Parameters: 13,000,000,000
Dataset size: 2T
Training hardware: NVIDIA A100 SXM4 80 GB
Training cost (2023 USD): $235K
Model accessibility: Open weights (restricted use)
Open weights: Yes
Citations: 16,911
Epoch confidence: Confident

Benchmark results

01	Epoch Capabilities Index	ECI Score	105.9
02	BoolQ	Score	0.82
03	PIQA	Score	0.81
04	HellaSwag	Overall accuracy	0.81
05	TriviaQA	EM	0.8
06	LAMBADA	Score	0.77
07	WinoGrande	Accuracy	0.73
08	ARC AI2	Challenge score	0.6
09	OpenBookQA	Accuracy	0.57
10	ScienceQA	Score	0.56
11	MMLU	EM	0.56