Meta
Llama 4 Maverick
Open multimodal deployments
text
image
Open weights
Input per 1M tokens
$0.270
Output per 1M tokens
$0.850
Context window
1M
Max output
—
Reasoning
8.4/10
Coding
8.3/10
Speed
180 tok/s
Overall
8.4/10
What it costs you a month
Change the numbers to match your workload. Cached input is billed lower where the maker offers it (— per million).
Monthly
$15.75
Yearly
$189
Cheaper option
Grok 4 Fast does the same job for $10.50 a month — saving $5.25.