Using Llama 4 Maverick 17B Instruct FP8 on GCP Vertex AI
Implementation guide · Llama 4 · AI at Meta
ServerlessOpen Weights
GCP Vertex AI exposes Llama 4 Maverick 17B Instruct FP8 through model ID llama-4-maverick-17b-128e-instruct-fp8. Use the setup steps, sourced pricing, capabilities, and official provider links below to validate this route before deployment.
Last refreshed 2026-07-09. Next refresh: weekly.
Quick Start
- 1
- 2Use the GCP Vertex AI SDK or REST API to call
llama-4-maverick-17b-128e-instruct-fp8— see the documentation for request format. - 3
Code Examples
Install
pip install google-cloud-aiplatformAPI key
GOOGLE_CLOUD_PROJECTModel ID
llama-4-maverick-17b-128e-instruct-fp8For Google-published models use the model name directly, e.g. "gemini-2.0-flash-001". For third-party publishers (Anthropic, Meta, etc.) use the full publisher path, e.g. "publishers/anthropic/models/claude-3-5-sonnet-v2@20241022".
import os
import vertexai
from vertexai.generative_models import GenerativeModel
# Reads GOOGLE_CLOUD_PROJECT from env; authenticates via Application Default Credentials
vertexai.init(project=os.environ["GOOGLE_CLOUD_PROJECT"], location="us-central1")
model = GenerativeModel("llama-4-maverick-17b-128e-instruct-fp8")
response = model.generate_content("Hello")
print(response.text)Pricing on GCP Vertex AI
| Type | Price (per 1M) |
|---|---|
| Input tokens | $0.35 |
| Output tokens | $1.15 |
Capabilities
VisionMultimodalStructured Outputs
About Llama 4 Maverick 17B Instruct FP8
Meta's Llama 4 Maverick 17B with 128 experts, FP8-optimized for cost-efficient inference. Supports native Model Router integration on Microsoft Foundry.
Model Specs
Released2025-04-05
Parameters400B (17B active)
Context1m
ArchitectureMixture of Experts
Knowledge cutoff2024-08