modelbenchmark.io

Llama-3.2-90B-Vision-Instruct

Meta · llama-3-2-90b-vision-instruct

Compare

Open Llama multimodal model for image understanding and text reasoning

Specification

most-agreed values

Context
128K
Max output
8K
Released
2024-09-25
Knowledge cutoff
2023-12
Retires
2026-06-13
Open weights
yes
Input
text, image
Output
text

Price

US dollars per million tokens · most-agreed

Input
free
Output
free
Cache read
$0.175
Cache write
$0.70

Available from 5 hosts

HostIn $/MOut $/MCache rdCache wrContextOutputRetires
nvidiafreefree128K8K
io-net · meta-llama$0.35$0.40$0.175$0.7016K4K
oci · oci$2.00$2.00128K4K
watsonx$2.00$2.00128K128K
azure_ai$2.04$2.04128K2K2026-06-13

More from Meta

most-hosted first