Crusoe APIcrusoe.ai ↗
Retrieve real-time GPU instance pricing from Crusoe Cloud. Get on-demand and spot rates for all listed GPU models via a single API endpoint.
What is the Crusoe API?
The Crusoe Cloud GPU Pricing API exposes all GPU instance offerings listed on crusoe.ai through a single get_gpu_pricing endpoint, returning up to dozens of offering entries across on-demand and spot tenors. Each response includes 6 structured fields per offering: GPU model name, tenor type, price per hour, and related metadata. Entries where pricing requires a sales conversation return a null price_per_hour value rather than omitting the row.
No input parameters required.
curl -X GET 'https://api.parse.bot/scraper/6f9eb001-201d-4e45-9ffd-47ff79f04b04/get_gpu_pricing' \ -H 'X-API-Key: $PARSE_API_KEY'
Typed, relational, agent-ready
A generated client with real types, enums, and the links between objects — the structure a flat JSON response can't carry. Autocompletes in your editor and reads cleanly to coding agents.
- Fully typed · autocompletes
- Objects link to objects
- Typed errors & pagination
Typed Python client. Set up the SDK in your uv project, then pull this API’s typed client:
uv add parse-sdk uv run parse init uv run parse add --marketplace crusoe-ai-api
uv run parse add --marketplace pulls a pinned snapshot of this canonical API — it won’t change underneath you. To customize it, subscribe and swap to your own copy.
"""Walkthrough: Crusoe Cloud GPU Pricing — list all offerings with prices."""
from parse_apis.crusoe_ai_api import CrusoeCloud, ParseError
client = CrusoeCloud()
# Fetch GPU offerings, capped at 20 total items
try:
for offering in client.offerings.list(limit=20):
print(f"{offering.gpu_model} ({offering.memory}) [{offering.tenor}]: {offering.price_display}")
except ParseError as e:
print(f"Extraction failed ({e.code}): {e}")
print("exercised: offerings.list")
Returns all GPU instance offerings from the Crusoe Cloud pricing page. Each GPU model is returned with both on-demand and spot tenor entries. Prices are per-GPU-hour where listed; some offerings show 'Contact sales' (price_per_hour will be null). The site does not expose gpu_count, region, or availability data in the pricing table, so these fields are always null.
No input parameters required.
{
"type": "object",
"fields": {
"total": "integer count of offering entries returned",
"offerings": "array of GPU offering objects with pricing details"
},
"sample": {
"data": {
"total": 18,
"offerings": [
{
"tenor": "on-demand",
"memory": "141GB",
"region": null,
"gpu_count": null,
"gpu_model": "NVIDIA H200 HGX",
"availability": null,
"is_from_price": false,
"price_display": "$4.29/GPU-hr",
"price_per_hour": 4.29
},
{
"tenor": "spot",
"memory": "141GB",
"region": null,
"gpu_count": null,
"gpu_model": "NVIDIA H200 HGX",
"availability": null,
"is_from_price": false,
"price_display": "Contact sales",
"price_per_hour": null
}
]
},
"status": "success"
}
}About the Crusoe API
What the API Returns
The get_gpu_pricing endpoint takes no input parameters and returns a total count alongside an offerings array. Each object in that array describes one GPU instance offering from Crusoe Cloud's pricing page. Fields include the GPU model identifier, the tenor (either on-demand or spot), and a price_per_hour value. When Crusoe lists a GPU as "Contact sales" rather than publishing a rate, price_per_hour is null — the offering still appears in the response so you have a complete picture of what's available.
Pricing Tiers Covered
Each GPU model typically appears twice in the offerings array: once for the on-demand tenor and once for the spot tenor. This lets you directly compare the two price points for any given GPU without making multiple calls. Spot entries tend to carry lower rates where Crusoe publishes them, but are also more likely to surface as null depending on availability status on the source page.
Scope and Limitations
The API covers GPU offerings as they appear on the Crusoe Cloud pricing page. It does not expose compute instance configurations such as vCPU count, RAM, or storage specs, nor does it include regional availability breakdowns or per-GPU-count pricing (Crusoe does not publish that granularity on its public page). The response reflects the current state of the pricing page at request time.
The Crusoe API is a managed, monitored endpoint for crusoe.ai — not a raw scraper you maintain. Every endpoint is automatically health-checked on a schedule, and when crusoe.ai changes and a check fails, the API is automatically queued for repair and re-verified. It is built to keep working as the site underneath it changes.
This isn't an official crusoe.ai API — it's an independent, maintained REST wrapper over public data. Where the source has no official API (or only a limited one), Parse gives you a stable contract over a source that never promised one, and keeps it current. Need a new endpoint or field? You can revise it yourself in plain English and the agent rebuilds it against the live site in minutes — contributing the change back to the shared API is free.
Will this API break when the source site changes?+
Is this an official API from the source site?+
Can I fix or extend this API myself if I need a new endpoint or field?+
What happens if I call an endpoint that has an issue?+
- Track Crusoe Cloud GPU spot prices over time to identify cost-effective windows for batch workloads
- Compare on-demand versus spot
price_per_hourfor each GPU model to decide which tenor fits a project's fault tolerance - Feed GPU price data into an internal cost-estimation tool alongside offerings from other cloud providers
- Flag when a previously null
price_per_hourentry (Contact sales) becomes a published rate - Build a dashboard displaying current Crusoe Cloud pricing for GPU procurement teams
- Alert when spot pricing for a specific GPU model drops below a defined threshold
| Tier | Price | Credits/month | Rate limit |
|---|---|---|---|
| Free | $0/mo | 200 | 5 req/min |
| Hobby | $30/mo | 1,000 | 20 req/min |
| Developer | $100/mo | 5,000 | 100 req/min |
| Team | $300/mo | 20,000 | 300 req/min |
| Company | $1,000/mo | 100,000 | 500 req/min |
Each endpoint has a fixed posted price per successful call — most fall between 1 and 10 credits — shown on this API's page before you run it. Exceeding the rate limit returns a 429 response. Authenticate with the X-API-Key header.
Does Crusoe Cloud have an official developer API?+
What does `price_per_hour` look like when Crusoe hasn't published a rate?+
null. The offering object still appears in the array with all other available fields populated, so you can see that the GPU model exists and which tenor it covers, even without a numeric rate.Does the API distinguish between different GPU counts per instance (e.g. 1x vs 8x H100)?+
gpu_count is not a field in the response. The API returns offerings at the model level. You can fork this API on Parse and revise it to add that field if Crusoe updates their page to expose it.