Llama 4 Maverick

Open-weights natively multimodal MoE (17B active, 128 experts). Pricing varies by host (Together, Groq, Fireworks). Open weights — API prices vary by hosting provider.

Llama 4 Maverick strengths

• Open weights / self-hostable
• Natively multimodal
• Strong price-performance
• Wide host availability

Pricing & context

Cost guide: a typical call of about 10K input + 2K output tokens costs roughly $0.003 at list prices. Worth modelling against cheaper tiers before committing high-volume traffic.

When to choose Llama 4 Maverick

Llama 4 Maverick is best for teams wanting an open, multimodal model they can self-host or run cheaply via multiple providers. If your workload is more cost-sensitive, weigh it against gpt-oss-120b (≈$0.03 input /1M) first.

Llama 4 Maverick FAQ

How much does Llama 4 Maverick cost?

Llama 4 Maverick is priced at ≈$0.15 per 1M input tokens and ≈$0.60 per 1M output tokens (public API list price), with a 1M tokens context window. A typical call of about 10K input and 2K output tokens costs roughly $0.003.

What is Llama 4 Maverick best for?

Llama 4 Maverick by Meta is best for teams wanting an open, multimodal model they can self-host or run cheaply via multiple providers.

How does Llama 4 Maverick pricing compare to Nemotron 3 Ultra?

Llama 4 Maverick input costs ≈$0.15 per 1M tokens versus $0.50 for Nemotron 3 Ultra, roughly 3.3x less expensive on input. Output is ≈$0.60 vs $2.20.

Is Llama 4 Maverick multimodal?

Llama 4 Maverick supports text, image.

Tools that use Llama 4 Maverick