Llama 4 Maverick: Benchmarks, Pricing & Size
Llama 4 Maverick is a language model from Meta, released in April 2025, with multimodal input.
Llama 4 Maverick is a natively multimodal model capable of processing both text and images. It features a 17 billion active parameter mixture-of-experts (MoE) architecture with 128 experts, supporting a wide range of multimodal tasks such
Llama 4 Maverick pricing
Providers
Llama 4 Maverick starts at $0.170 per million input tokens and $0.600 per million output tokens via DeepInfra. See all 7 providers below with their per-token pricing, latency, throughput, and modality support.
| Provider | Input $/M | Output $/M | Max Input | Max Output | Latency p95 s | Throughput P95 | Quant | Input | Output |
|---|---|---|---|---|---|---|---|---|---|
 DeepInfra | $0.170 | $0.600 | 1.0M | 1.0M | 0.38 | 84 c/s | — | ||
 Novita | $0.170 | $0.850 | 1.0M | 1.0M | 0.62 | 69 c/s | — | ||
 Lambda | $0.180 | $0.600 | 1.0M | 1.0M | 0.65 | 94 c/s | — | ||
 Groq | $0.200 | $0.600 | 1.0M | 1.0M | 0.27 | 307 c/s | — | ||
 Fireworks | $0.220 | $0.880 | 1.0M | 1.0M | 0.62 | 63 c/s | — | ||
 Together | $0.270 | $0.850 | 1.0M | 1.0M | 0.20 | 98 c/s | — | ||
 Sambanova | $0.630 | $1.79 | 1.0M | 1.0M | 2.04 | 639 c/s | — |
Llama 4 Maverick model size
Llama 4 Maverick has 400 billion parameters and was trained on 22 trillion tokens. See how it compares to other models in the same parameter range.
Llama 4 Maverick API
API access coming soon
Llama 4 Maverick will be available through our gateway shortly.
Llama 4 Maverick examples
Recent arena outputs from Llama 4 Maverick, picked from the highest-ranked matchups.
Llama 4 Maverick license
Llama 4 Maverick is released under the Llama 4 Community License Agreement license, which restricts commercial use, has 400.0B parameters.
- License
- Llama 4 Community License Agreement
- Non-commercial
- Parameters
- 400.0B
FAQ
Common questions about Llama 4 Maverick.