{T}
TokenMath.net
Official Provider HubMenlo Park, CA, USA

Meta Llama API Pricing & Token Economics

Meta's Llama family (Llama 4 Scout, Llama 3.3 70B, and Llama 3.1 405B) represents the premier open-weights architecture in AI. Hosted by serverless cloud partners such as Together AI, Groq, and AWS Bedrock, Llama models provide ultra-competitive per-token rates without proprietary vendor lock-in.

Prompt Caching Policy

Batch / Asynchronous Rules

Standard API execution rules. Refer to individual endpoint rate limits for asynchronous processing.

Special Billing Conditions

Zero write fees on standard prefix evaluation; automatic cache invalidation.

Active Meta Llama Models(3 models verified)

All 30 Models →
ModelContextInput / 1MCached / 1MOutput / 1MBatch DiscountActions
Llama 4 Scout (17B Active / 109B MoE) — Together AI
meta-llama/Llama-4-Scout-17B-16E-Instruct
328k$0.18$0.59N/ASpecs →
Llama 4 Maverick (400B) — Together AI
meta-llama/llama-4-maverick-instruct
1M$0.80$1.60N/ASpecs →
Llama 3.1 405B — Together AI
meta-llama/Meta-Llama-3.1-405B-Instruct-Turbo
128k$3.50$3.50N/ASpecs →

Calculate your monthly Meta Llama bill

Simulate prompt caching hit rates, batch discounts, and output token costs with zero server logging.

Open Cost Calculator →