Meta / Groq · Ultra Fast · Updated 2026-11

Llama 3.3 70B (Groq) API Pricing & Cost Breakdown

Ultra-low-latency streaming UX, real-time voice agents, and open-weights portability.

Input / 1M Tokens
$0.59
Cached Input / 1M
$0.590
Output / 1M Tokens
$0.79
Context Window
128K

Workload Cost Scenarios (30-Day Billing)

Open-weights 70B model served on Groq LPU at high token/sec throughput.

Scenario In / Out per Req Uncached Monthly 50% Cache Hit Monthly
Hobby / Prototype (100 req/day) 1500 / 400 $3.60 $3.60
Growing Startup (2,500 req/day) 2500 / 700 $152.10 $152.10
Production Scale (50,000 req/day) 3000 / 800 $3,603.00 $3,603.00

Compare Llama 3.3 70B (Groq) Against Alternatives

Lower-Cost Alternative Models

Gemini 2.0 Flash ($0.40/1M out) Qwen 2.5 72B ($0.40/1M out) Grok 3 Mini ($0.50/1M out) GPT-4o mini ($0.60/1M out)

Frequently Asked Questions

What is Llama 3.3 70B (Groq) API pricing per 1M tokens?

Standard input is $0.59/1M, cached input is $0.590/1M, and output is $0.79/1M.

What is the context window of Llama 3.3 70B (Groq)?

Llama 3.3 70B (Groq) supports a 128K token context window with up to 32,768 max output tokens.

How much does prompt caching save on Llama 3.3 70B (Groq)?

Prompt caching reduces cached input token rates by 0% compared with uncached input.

What workloads is Llama 3.3 70B (Groq) best suited for?

Ultra-low-latency streaming UX, real-time voice agents, and open-weights portability.