Rate limits
SambaNova Cloud enforces rate limits on inference requests per model to ensure that developers are able to try the fastest inference.
Developer tier
Rate limits for the developer tier are described below.
Preview models
Preview models in SambaNova Cloud are offered as early access models primarily for trial purposes. During the Preview phase, these models are available exclusively in the playground.
For API access and higher rate limits for DeepSeek-R1, please complete this form to join the waitlist. Priority will be given to developers who are actively using SambaNova Cloud and have signed up with a payment method.
For those who have been granted access for DeepSeek-R1 from the waitlist, please refer to the Production models section under the Payment and credits tab for the allowed rate limits.
Developer | Model ID | Requests per minute (RPM) | Requests per hour (RPH) | Requests per day (RPD) |
---|---|---|---|---|
DeepSeek | ||||
DeepSeek-R1 | 2 | 5 | 10 |
Production models
Production models meet our high standards for speed and quality and are intended for use in production environments.
Payment and credits limits are applied when a payment method is linked with the account. Credits only limits are applied when there is no additional payment method linked with the account. See more on the Billing page.
Developer | Model ID | Requests per minute (RPM) | Requests per hour (RPH) | Requests per day (RPD) |
---|---|---|---|---|
DeepSeek | ||||
DeepSeek-R1 | 10 | 100 | 2000 | |
DeepSeek-R1-Distill-Llama-70B | 30 | 300 | 3600 | |
Ai2 | ||||
Llama-3.1-Tulu-3-405B | 15 | 150 | 1800 | |
Meta | ||||
Meta-Llama-3.3-70B-Instruct | 30 | 300 | 3600 | |
Meta-Llama-3.2-3B-Instruct | 60 | 600 | 7200 | |
Meta-Llama-3.2-1B-Instruct | 60 | 600 | 7200 | |
Meta-Llama-3.1-405B-Instruct | 15 | 150 | 1800 | |
Meta-Llama-3.1-70B-Instruct | 30 | 300 | 3600 | |
Meta-Llama-3.1-8B-Instruct | 60 | 600 | 7200 | |
Meta-Llama-Guard-3-8B | 30 | 300 | 3600 | |
Llama-3.2-90B-Vision-Instruct | 5 | 50 | 600 | |
Llama-3.2-11B-Vision-Instruct | 15 | 150 | 1800 | |
Qwen | ||||
Qwen2.5-72B-Instruct | 20 | 200 | 2400 | |
Qwen2.5-Coder-32B-Instruct | 20 | 200 | 2400 | |
QwQ-32B | 10 | 100 | 1200 | |
Qwen2-Audio-7B-Instruct | 5 | 50 | 600 | |
Tokyotech-llm | ||||
Llama-3.1-Swallow-8B-Instruct-v0.3 | 30 | 300 | 3600 | |
Llama-3.1-Swallow-70B-Instruct-v0.3 | 20 | 200 | 2400 |
Other tiers
For rate limits in the Managed Subscription and Dedicated tiers, reach out to sales or contact us on our Community page so we can accommodate your projects needs.