Skip to main content
Meta/ Llama

Llama 4 Maverick

meta/llama/4-maverick

Llama 4 Maverick 17B Instruct (128E) is a high-capacity multimodal language model from Meta, built on a mixture-of-experts (MoE) architecture with 128 experts and 17 billion active parameters per forward...

Released
2025-04-05
Context window
1M
Max output
16K
Weights
Open weight

Modalities

Input
TextImage
Output
Text

Identity

Version
llama-4-maverick
Knowledge cutoff
2024-08-31

Resources

Providers

ProviderProvider model IDProtocolContextMax outputInputOutputCapabilitiesNotesPricingStreaming
OpenRoutermeta-llama/llama-4-maverickopenai_chat1M16K
TextImage
Text
Tool callStructuredTemperatureOpen weight
Hugging Face: meta-llama/Llama-4-Maverick-17B-128E-Instruct
input $0.2/M
output $0.8/M
Yes
NanoGPTmeta-llama/llama-4-maverickopenai_chat1M16K
TextImage
Text
Tool callStructuredTemperatureOpen weight
-
input $0.15/M · cache read $0.075/M
output $0.6/M
Yes
Kilo Gatewaymeta-llama/llama-4-maverickopenai_chat1M16K
TextImage
Text
Tool callStructuredTemperatureOpen weight
-
input $0.2/M
output $0.8/M
Yes

Capabilities

Tool callStructuredTemperatureOpen weight

FAQ