modelbenchmark.io

Llama 3.2 3b Instruct

Meta · llama-3-2-3b-instruct

Compare

Open Llama multimodal model for image understanding and text reasoning

Specification

most-agreed values

Context
131K
Max output
8K
Released
2024-01-01
Knowledge cutoff
2023-12-31
Retires
Open weights
yes
Input
text, pdf
Output
text

Price

US dollars per million tokens · most-agreed

Input
$0.0306
Output
$0.0493
Cache read
$0.0153
Cache write
$0.10

Rate limits

Requests per minute
300

Available from 17 hosts

HostIn $/MOut $/MCache rdCache wrContextOutputRetires
nvidiafreefree33K32K
deepinfra$0.02$0.02131K131K
inference$0.02$0.0216K4K
llmgateway$0.03$0.0533K32K
llmgateway-providers$0.03$0.0533K32K
novita$0.03$0.0533K32K
novita-ai · meta-llama$0.03$0.0533K32K
nano-gpt · meta-llama$0.0306$0.0493$0.0153131K8K
kilo · meta-llama$0.05$0.33131K118K
openrouter$0.05$0.33131K118K
openrouter · meta-llama$0.05$0.33131K118K
cloudflare$0.0509$0.33580K80K
cloudflare-workers-ai · cf$0.0509$0.33580K80K
together_ai$0.06$0.06
pioneer · meta-llama$0.10$0.335$0.10$0.10131K80K
hyperbolic$0.12$0.3033K33K
watsonx$0.15$0.15128K128K

Other listings of this model

same model, different key

More from Meta

most-hosted first