​
Availability

Current region availability:

RegionLaunch statusHardware availability
US_ILLINOIS_2Generally AvailableNVIDIA_A100_80GB
US_TEXAS_1Generally AvailableNVIDIA_H100_80GB
US_VIRGINIA_2Generally AvailableAMD_MI300X_192GB
AP_TOKYO_1Enterprise onlyNVIDIA_H100_80GB
EU_FRANKFURT_1Enterprise onlyNVIDIA_H100_80GB
EU_LONDON_1Enterprise onlyNVIDIA_H100_80GB
US_ILLINOIS_1Enterprise onlyNVIDIA_H100_80GB
US_IOWA_1Enterprise onlyNVIDIA_H100_80GB
US_VIRGINIA_1Enterprise onlyNVIDIA_H100_80GB

If you need deployments in a non-GA region, please contact our team at inquiries@fireworks.ai.

​
Using a region

When creating a deployment, you can pass the --region flag:

firectl create deployment accounts/fireworks/models/llama-v3p1-8b-instruct \
    --region US_IOWA_1

​
Changing regions

Updating a region for a deployment in-place is currently not supported. To move a deployment between regions, please create a new deployment in the new region, then delete the old deployment.

​
Quotas

Each region has it’s own separate quota for each hardware type. To view your current quotas, run

firectl list quotas

Was this page helpful?

On-demand deploymentsReserved capacity