← Back to all pricing

Llama 3.3 70B API Pricing — $0.59/M input, $0.79/M output

Stable performance: deep knowledge with high reliability.

Groq decommission Aug 16, 2026 — use Muse Spark 1.1 (Meta)
Announced 2026-06-17. Shuts down 2026-08-16. Source · Migration guide · Full retirement tracker
Input
$0.59/M
Output
$0.79/M
Blended
$0.64/M
Provider
Verified 2026-04-06 — source

Speed

Not yet measured — see the speed benchmark leaderboard for models we do track.

Cost at scale

Tokens / monthEst. cost (blended 3:1)
100,000$0.06
1,000,000$0.64
10,000,000$6.40
100,000,000$64.00

How Llama 3.3 70B compares

Llama 3.1 8B — $0.06/MGPT-OSS 20B — $0.13/MGPT-OSS 120B — $0.26/MLlama 4 Maverick — $0.30/MQwen 3.8 30B — $1.20/MGPT-5 Mini — $0.69/MGemini 3.5 Flash Lite — $0.56/MGemini 3.1 Flash Lite — $0.56/M
See all Groq models →

Head-to-head comparisons

Llama 3.3 70B vs GPT-OSS 120B

FAQ

Is Llama 3.3 70B cheaper than GPT-5 Mini?

Llama 3.3 70B costs $0.64/M blended tokens, GPT-5 Mini costs $0.69/M — Llama 3.3 70B is cheaper.

How much does 1 million tokens cost with Llama 3.3 70B?

At a 3:1 input:output ratio, 1 million blended tokens costs approximately $0.64. Pure input costs $0.59/M; pure output costs $0.79/M.

Who provides Llama 3.3 70B?

Llama 3.3 70B is served by Groq. Pricing was last verified on 2026-04-06 against https://console.groq.com/docs/models.

Try Llama 3.3 70B for free

Run real prompts against Llama 3.3 70B and every other model on this page in one workspace.

Try Llama 3.3 70B Free