{
  "opencollection": "1.0.0",
  "info": {
    "name": "ScalarLM Generate API",
    "version": "1.151"
  },
  "items": [
    {
      "info": {
        "name": "Generate",
        "type": "folder"
      },
      "items": [
        {
          "info": {
            "name": "Submit a batch of prompts to the inference queue",
            "type": "http"
          },
          "http": {
            "method": "POST",
            "url": "https://gpt-oss.cray-lm.com/v1/generate",
            "body": {
              "type": "json",
              "data": "{}"
            }
          },
          "docs": "Enqueues a batch of independent inference requests. Each prompt entry is a bare string, a {prompt: ...} object, or a {messages: [...]} object rendered with the model chat template at enqueue time."
        },
        {
          "info": {
            "name": "Fetch results for queued request ids",
            "type": "http"
          },
          "http": {
            "method": "POST",
            "url": "https://gpt-oss.cray-lm.com/v1/generate/get_results",
            "body": {
              "type": "json",
              "data": "{}"
            }
          },
          "docs": "Fetch results for queued request ids"
        },
        {
          "info": {
            "name": "Pull work items from the inference queue",
            "type": "http"
          },
          "http": {
            "method": "POST",
            "url": "https://gpt-oss.cray-lm.com/v1/generate/get_work",
            "body": {
              "type": "json",
              "data": "{}"
            }
          },
          "docs": "Worker-facing: claims up to batch_size queued requests plus any new adaptors."
        },
        {
          "info": {
            "name": "Return completed work to the queue",
            "type": "http"
          },
          "http": {
            "method": "POST",
            "url": "https://gpt-oss.cray-lm.com/v1/generate/finish_work",
            "body": {
              "type": "json",
              "data": "{}"
            }
          },
          "docs": "Worker-facing: acknowledges one or more claimed requests with their response, error and token/flop accounting."
        },
        {
          "info": {
            "name": "Upload inference payload data",
            "type": "http"
          },
          "http": {
            "method": "POST",
            "url": "https://gpt-oss.cray-lm.com/v1/generate/upload"
          },
          "docs": "Upload inference payload data"
        },
        {
          "info": {
            "name": "Download the payload for a request id",
            "type": "http"
          },
          "http": {
            "method": "POST",
            "url": "https://gpt-oss.cray-lm.com/v1/generate/download",
            "body": {
              "type": "json",
              "data": "{}"
            }
          },
          "docs": "Download the payload for a request id"
        },
        {
          "info": {
            "name": "Clear the inference work queue",
            "type": "http"
          },
          "http": {
            "method": "POST",
            "url": "https://gpt-oss.cray-lm.com/v1/generate/clear_queue"
          },
          "docs": "Clear the inference work queue"
        },
        {
          "info": {
            "name": "List adaptors newer than the caller's loaded set",
            "type": "http"
          },
          "http": {
            "method": "POST",
            "url": "https://gpt-oss.cray-lm.com/v1/generate/get_adaptors",
            "body": {
              "type": "json",
              "data": "{}"
            }
          },
          "docs": "List adaptors newer than the caller's loaded set"
        },
        {
          "info": {
            "name": "Inference queue metrics",
            "type": "http"
          },
          "http": {
            "method": "GET",
            "url": "https://gpt-oss.cray-lm.com/v1/generate/metrics"
          },
          "docs": "Inference queue metrics"
        },
        {
          "info": {
            "name": "List recent inference requests",
            "type": "http"
          },
          "http": {
            "method": "GET",
            "url": "https://gpt-oss.cray-lm.com/v1/generate/list_requests",
            "params": [
              {
                "name": "cursor",
                "value": "",
                "type": "query",
                "description": "Opaque timestamp cursor for pagination."
              },
              {
                "name": "limit",
                "value": "",
                "type": "query",
                "description": "Page size."
              }
            ]
          },
          "docs": "List recent inference requests"
        },
        {
          "info": {
            "name": "Get one inference request",
            "type": "http"
          },
          "http": {
            "method": "GET",
            "url": "https://gpt-oss.cray-lm.com/v1/generate/request/:request_id",
            "params": [
              {
                "name": "request_id",
                "value": "",
                "type": "path"
              }
            ]
          },
          "docs": "Get one inference request"
        },
        {
          "info": {
            "name": "List routes registered on the generate router",
            "type": "http"
          },
          "http": {
            "method": "GET",
            "url": "https://gpt-oss.cray-lm.com/v1/generate/endpoints"
          },
          "docs": "List routes registered on the generate router"
        }
      ]
    }
  ],
  "bundled": true
}