Skip to content
Model prices

Vertex AI (Llama) API pricing

11 Vertex AI (Llama) models, in US dollars per million tokens. Updated Oct 28, 2026.

11 models
US dollars per million tokens
ModelInputCached inputOutputContextNotes
meta/llama-3.1-405b-instruct-maas$5.00—$16.00128K
meta/llama-3.1-70b-instruct-maas$0—$0128K
meta/llama-3.1-8b-instruct-maas$0—$0128K
meta/llama-3.2-90b-vision-instruct-maas$0—$0128K
meta/llama-4-maverick-17b-128e-instruct-maas$0.35—$1.151M
meta/llama-4-maverick-17b-16e-instruct-maas$0.35—$1.151M
meta/llama-4-scout-17b-128e-instruct-maas$0.25—$0.7010M
meta/llama-4-scout-17b-16e-instruct-maas$0.25—$0.7010M
meta/llama3-405b-instruct-maas$0—$032K
meta/llama3-70b-instruct-maas$0—$032K
meta/llama3-8b-instruct-maas$0—$032K

See what your Vertex AI (Llama) calls actually cost, by service, model and customer, priced for where each call ran.

Get started free

List prices from LiteLLM's public price file. Long-prompt, batch and priority rates aren't shown; check Vertex AI (Llama)'s pricing page before you commit to a budget. All providers