Skip to content
ACS Infra

API reference

The reference comes from the API's OpenAPI spec, so it changes when the API changes. Each route's page lists the request fields and each response, and its playground sends the request to the API.

RouteWhat it does
GET /v1/modelsList the Base models.
GET /v1/models/{model_id}Read one Base model.
GET /v1/models/{model_id}/statusSay whether a model is Warm.
POST /v1/completionsCreate a Completion.
POST /v1/completion-jobsQueue a Job.
GET /v1/completion-jobs/{job_id}Read a Job.
GET /v1/completion-jobs/{job_id}/resultFetch the result of a Job.
POST /v1/completion-jobs/{job_id}/cancelCancel a Job.
POST /v1/harvestQueue a harvest Job of Activations.
GET /v1/harvest/{job_id}Read a harvest Job and its files, and wait for its status to change.
DELETE /v1/harvest/{job_id}Cancel a harvest Job.
GET /v1/loomsList your Looms, pinned first.
POST /v1/loomsCreate a Loom.
POST /v1/looms/importMake a Loom from a file.
GET /v1/looms/{loom_id}Read a Loom.
POST /v1/looms/{loom_id}Rename or pin a Loom.
GET /v1/looms/{loom_id}/exportDownload a Loom as a file.
GET /v1/looms/{loom_id}/streamFollow a Loom as Server-Sent Events.
POST /v1/looms/{loom_id}/cancelStop the Loom's growing Nodes.
POST /v1/looms/{loom_id}/rootReplace the tree with a new root.
DELETE /v1/looms/{loom_id}Delete a Loom.
POST /v1/looms/{loom_id}/nodes/{node_id}Edit a Node.
POST /v1/looms/{loom_id}/nodes/{node_id}/splitSplit a Node at an offset.
POST /v1/looms/{loom_id}/nodes/{node_id}/siblingWrite a Node beside a Node.
POST /v1/looms/{loom_id}/nodes/{node_id}/branchBranch from a Node of a Loom.
DELETE /v1/looms/{loom_id}/nodes/{node_id}Delete a Node and the Nodes under it.
GET /healthCheck that the API answers.
POST /v1/chat/completionsRefuse every chat request with 400 and a message that names POST /v1/completions.

List Models GET

List the Base models that the API serves, each with its Model id, the Checkpoint it serves, its GPUs, and its capabilities: its context length, Layers, hidden size, and the limits of a request to it. `default` marks the model of a request without `model`. The API lists them from its settings and asks no Inference, so a list never starts a GPU.