| title | cloudfit |
|---|---|
| emoji | 🖥️ |
| colorFrom | blue |
| colorTo | green |
| sdk | gradio |
| sdk_version | 5.49.1 |
| app_file | app.py |
| pinned | false |
| license | apache-2.0 |
| short_description | Pick a workload, get a cloud machine-type recommendation. |
Gradio demo for cloudfit-core. Pick a workload, get a ranked list of GCP machine types scored on cost, performance, and availability.
Live demo: chaitanyakasaraneni-cloudfit-ui.hf.space
A one-screen UI that runs the cloudfit scoring engine in-process against a bundled GCP machine-type snapshot. No login, no credentials, no provider calls at request time. It exists so you can try cloudfit in 30 seconds before reading any code.
For programmatic access, use the HTTP API instead.
pip install -r requirements.txt
python app.py
# open http://127.0.0.1:7860data/gcp_snapshot.json is a representative sample of GCP Compute Engine across us-central1, us-east1, us-west1, europe-west4, and asia-southeast1. 875 entries total. Prices are adjusted with approximate per-region multipliers. Not live pricing: regenerate with cloudfit-provider-gcp if you need fresher numbers.
cloudfit-core: scoring enginecloudfit-api: HTTP APIcloudfit-provider-gcp: GCP machine-type fetchercloudfit-provider-aws: AWS provider (planning phase)
Apache 2.0: see LICENSE.