How much does Llama3.1.8b cost?

Llama3.1.8b costs $0.10 per 1M input tokens and $0.10 per 1M output tokens. Prices are sourced from official Cerebras Systems documentation.

Is Llama3.1.8b available via API?

Yes. Llama3.1.8b is available via the Cerebras API. Tickerr monitors its uptime and latency every 5 minutes.

Operational· provider status page

Input

$0.10

Output

$0.10

Cached input

—

Batch input

—

Effective from Apr 30, 2026. Source: official provider documentation.

Estimated monthly cost · 70% input / 30% output split

cheapestLlama3.1.8b$0.100

Gpt Oss 120b$0.470

Qwen 3.32b$0.520

Llama3.1.70b$0.600

Llama 3.3.70b$0.955

Zai Glm 4.6$2.40