Llama 2 Chat 7B

Last updated: Recently

Model Type

Text

Max Context

Max Output

Parameter Scale

Llama 2 Chat 7B by Meta: Text model; TTFT 0.973s, 121.3 tok/s.

Input Modality

Output Modality

Inference Speed

121.350 tokens/s

Latest Release Date

7/18/2023

SDK Ecosystem

artificial-analysis

Model Overview Fields

name

Llama 2 Chat 7B

Release date

7/18/2023

Performance

121.350 tokens/s

model_creator.name

Meta

AA Evaluation Scores

AA Intelligence Index

9.7

AA Coding Index

AA Math Index

MMLU Pro

0.164

GPQA

0.227

HLE

0.058

LiveCodeBench

0.002

SciCode

Math-500

0.059

AIME

AIME 2025

IFBench

LCR

TerminalBench Hard

TAU2

Output Speed (tokens/s)

121.35 tok/s

TTFT (s)

0.973s

First Answer Token (s)

0.973s