Skip to main content
POST
Place a bid
CLI equivalent, a market bid for nodes: ornn market bid --gpu-slug <gpu-slug> --nodes 2 --yes. Pass --idempotency-key to make a retry safe. Send a new Idempotency-Key for each bid. If the response is lost, retry with the same key and body: you get the original order back instead of a second bid. See Rent GPUs on Spot.

Authorizations

Authorization
string
header
required

Create an organization API key in Account Settings > Developer. Do not send this key to the Kubernetes API server.

Headers

Idempotency-Key
string
required

Required. A unique key of 1 to 200 characters for this operation, such as a UUID. Retrying with the same key and the same body returns the original result with HTTP 200 instead of creating a second one. The key is remembered for 24 hours per organization; reusing it for a different request returns 409 idempotency_conflict. A missing key returns 400 idempotency_key_required.

Required string length: 1 - 200

Body

application/json

Send exactly one of gpus or nodes; nodes needs orderType market.

gpuSlug
string
required

Catalog GPU slug of the order book.

orderType
enum<string>
required

limit takes capacity at or below your price and rests the rest until it fills or you cancel. market takes what is for sale now and cancels the rest.

Available options:
limit,
market
saleType
enum<string>
default:spot

spot: a reservation on the node can end the VM. on-demand: only a stop or the listing's end ends it.

Available options:
spot,
on-demand
region
string

A region with an order book, for example us-east-11 (see GET /v1/markets/books). Omit to fill at any location. A region without a book is refused with unknown_region.

gpus
integer

GPUs, taken from the lowest asks first; one node can supply part of them and a bid can span nodes. Send this or nodes.

Required range: x >= 1
nodes
integer

Distinct nodes with a GPU free, each giving every GPU free on it. Market orders only (nodes_need_market_order on a limit). Send this or gpus.

Required range: x >= 1
pricePerGpuHourMicroUsd
integer<int64>

Required for limit bids; omit for market bids. The most you pay per GPU-hour, fees included, in integer micro-USD (1000000 is $1.00).

Required range: x >= 1
diskId
string<uuid>

The disk every VM boots, from your disks or Ornn's. Omit to boot your newest disk, else Ornn's newest.

Response

A retry with the same Idempotency-Key and body; returns the original order's current state.

order
object
required