Quotas
Quotas are usage limits placed on your account. Most quotas are not configurable.
Default quotas
By default, the following quotas are in place:
Quota Name | Default Value | Can be raised? |
---|---|---|
Serverless inference RPM | 600 | No |
# of deployed models | 100 | Yes |
# A100 GPUs | 8 | Yes |
# H100 GPUs | 8 | Yes |
Monthly spend USD | $50 | Yes |
Viewing quotas
You can view your current quota capacity by running:
firectl list quotas
Raising quotas
Number of deployed models
This is available for enterprise accounts. Contact the Fireworks team at inquiries@fireworks.ai.
GPU quotas
GPU quotas are in place to limit the number of on-demand GPUs you can run. In order to raise your GPU quotas, you must purchase a reservation. Contact the Fireworks team at inquiries@fireworks.ai to purchase a reservation or to learn more.
Monthly spend
In order to prevent fraud, Fireworks imposes a monthly spending limit on your account. Once you hit the spending limit, your account will automatically enter a suspended state and all Fireworks usage will be stopped. This incldues serverless inference, dedicated deployments, and fine-tuning jobs.
Your spending limit will organically increase over time as you spend more on the platform. See the following table:
Tier | Spending Limit | Qualification |
---|---|---|
Tier 1 | $50/mo | Valid payment method added |
Tier 2 | $500/mo | Total historical spend of $100+ |
Tier 3 | $5,000/mo | Total historical spend of $1,000+ |
Tier 4 | $50,000/mo | Total historical spend of $10,000+ |
Unlimited | Unlimited | Contact us at inquiries@fireworks.ai |
You can purchase prepaid credits in order to move into the next tier.
Was this page helpful?