Skip to content
Sovyron

Llama 3.1 70B Instruct

llama131k context16k max outputreleased Jul 23, 2024Open weightsTool calling

As of Oct 4, 2026, the cheapest Llama 3.1 70B Instruct API price is $0.4 per 1M input tokens (Kilo Gateway) and $0.4 per 1M output tokens (Kilo Gateway). Llama 3.1 70B Instruct is offered by 6 providers in the llama family, open weights.

Cheapest input
$0.4
Cheapest output
$0.4
Official
—
output per 1M tokens
Providers
6
offering this model
Spread
1.8×
max / min output
Free offers
0
free-tier offers

API pricing by provider

Metered per-token prices. Cheapest input and output are highlighted; † marks a context-tier surcharge. How prices are ranked.

Llama 3.1 70B Instruct API prices by provider in USD per 1M tokens, updated Oct 4, 2026
Kilo GatewayopenCheapest$0.4Cheapest$0.4——131k—
OpenRouteropen$0.4$0.4——131k12d agoout −44.4%
Amazon Bedrockopen$0.72$0.72——128k—
Amazon Bedrockusopen$0.72$0.72——128k—
DevPass (LLM Gateway)open$0.72$0.72——128k—
LLM Gateway$0.72$0.72——128k—
Vercel AI Gateway$0.72$0.72——128k—

Llama 3.1 70B Instruct pricing FAQ

What is the cheapest Llama 3.1 70B Instruct API?

As of Oct 4, 2026, Kilo Gateway has the lowest Llama 3.1 70B Instruct output price at $0.4 per 1M tokens, and Kilo Gateway has the lowest input price at $0.4 per 1M tokens. Prices are metered per-token rates in USD; free tiers and subscription plans are excluded.

How many providers offer Llama 3.1 70B Instruct?

6 providers list Llama 3.1 70B Instruct on Sovyron; 7 of them sell it at a metered per-token price.

Is Llama 3.1 70B Instruct free?

No provider in the Sovyron catalog lists a free tier for Llama 3.1 70B Instruct.

What is the context window of Llama 3.1 70B Instruct?

Llama 3.1 70B Instruct supports a 131k-token context window and up to 16k output tokens.

Token cost calculator

What a month costs at one provider's published rates. Enter the tokens you expect to send and receive; cached tokens use the cached-input price where the provider lists one, and count as ordinary input where it does not.

Estimated

—