Inference API

Response Body

`application/json`

curl -X GET "https://example.com/v1/auth/me"

{
  "username": "string",
  "role": "string"
}

Response Body

`application/json`

curl -X GET "https://example.com/v1/responses/497f6eca-6276-4993-bfeb-53cbbbba6f08"

{
  "id": "string",
  "object": "response",
  "created_at": 0,
  "model": "string",
  "status": "in_progress",
  "incomplete_details": {
    "reason": "max_output_tokens"
  },
  "output": [
    {
      "type": "message",
      "id": "string",
      "role": "assistant",
      "status": "completed",
      "content": [
        {
          "type": "output_text",
          "text": "string",
          "annotations": [
            null
          ]
        }
      ]
    }
  ],
  "error": {
    "code": "string",
    "message": "string"
  },
  "usage": {
    "input_tokens": 0,
    "output_tokens": 0,
    "total_tokens": 0,
    "input_tokens_details": {
      "cached_tokens": 0
    },
    "output_tokens_details": {
      "reasoning_tokens": 0
    }
  },
  "service_tier": "default"
}

{
  "status_code": 400,
  "detail": "Bad Request",
  "extra": {}
}

Response Body

`application/json`

curl -X DELETE "https://example.com/v1/responses/497f6eca-6276-4993-bfeb-53cbbbba6f08"

{
  "id": "string",
  "object": "response",
  "created_at": 0,
  "model": "string",
  "status": "in_progress",
  "incomplete_details": {
    "reason": "max_output_tokens"
  },
  "output": [
    {
      "type": "message",
      "id": "string",
      "role": "assistant",
      "status": "completed",
      "content": [
        {
          "type": "output_text",
          "text": "string",
          "annotations": [
            null
          ]
        }
      ]
    }
  ],
  "error": {
    "code": "string",
    "message": "string"
  },
  "usage": {
    "input_tokens": 0,
    "output_tokens": 0,
    "total_tokens": 0,
    "input_tokens_details": {
      "cached_tokens": 0
    },
    "output_tokens_details": {
      "reasoning_tokens": 0
    }
  },
  "service_tier": "default"
}

{
  "status_code": 400,
  "detail": "Bad Request",
  "extra": {}
}

Query Parameters

limit?integer

Default100

stream_id?string|null

Response Body

`application/json`

curl -X GET "https://example.com/v1/responses"

{
  "object": "list",
  "data": [
    {
      "id": "string",
      "object": "response",
      "created_at": 0,
      "model": "string",
      "status": "in_progress"
    }
  ],
  "has_more": false
}

{
  "status_code": 400,
  "detail": "Bad Request",
  "extra": {}
}

Request Body

application/json

TypeScript Definitions

Use the request body type in TypeScript.

model*string

Model identifier, e.g. "zai-org/GLM-5"

input*|array<>

Input for the completion. Either a plain string (shorthand for a single user message) or an array of OpenAI Responses input items.

instructions?|

Developer instructions (system prompt).

tools?array<>|null

tool_choice?|string|null

max_output_tokens?integer|null

temperature?|

Rangevalue <= 2

top_p?|

Rangevalue <= 1

include_reasoning?boolean

Defaulttrue

enable_thinking?boolean|null

reasoning?|null

include?array<string>|

Ignored. Every value OpenAI defines here (web search, code interpreter, computer use, file search, logprobs, encrypted reasoning) depends on a feature this API does not implement, so the requested extra data is never present in the response regardless. Silently ignored rather than rejected: upstream routers such as requesty.ai count a 400 as a failed request against our reliability, while an ignored include degrades gracefully — the caller simply gets none of the extra data they would not have received anyway.

background?boolean

Defaultfalse

stream?boolean

Defaultfalse

text?|null

prompt_cache_key?|

Stable identifier for a sequence of requests that share a long prefix (a chat session, an agent loop). Requests with the same key are routed to the same worker while it stays warm, so the shared prefix is served from its prompt cache. Without it, placement is inferred from the input text.

service_tier?|

Processing tier for the request. "flex" selects flex processing (beta) on flex-enabled accounts: discounted tokens in exchange for lower scheduling priority and best-effort completion — the request may queue significantly longer than default-tier requests, so raise your client timeout accordingly, or use background: true and poll. All other accepted values (auto, default, scale, priority) are served at the default tier. Responses echo the tier actually served in service_tier.

metadata?

Extra metadata including "completion_window" ("24h" is the only supported value; requires background=true) and optional "stream_id" for grouping.

Response Body

`application/json`

curl -X POST "https://example.com/v1/responses" \  -H "Content-Type: application/json" \  -d '{    "model": "string",    "input": "string"  }'

{
  "id": "string",
  "object": "response",
  "created_at": 0,
  "model": "string",
  "status": "in_progress",
  "incomplete_details": {
    "reason": "max_output_tokens"
  },
  "output": [
    {
      "type": "message",
      "id": "string",
      "role": "assistant",
      "status": "completed",
      "content": [
        {
          "type": "output_text",
          "text": "string",
          "annotations": [
            null
          ]
        }
      ]
    }
  ],
  "error": {
    "code": "string",
    "message": "string"
  },
  "usage": {
    "input_tokens": 0,
    "output_tokens": 0,
    "total_tokens": 0,
    "input_tokens_details": {
      "cached_tokens": 0
    },
    "output_tokens_details": {
      "reasoning_tokens": 0
    }
  },
  "service_tier": "default"
}

{
  "status_code": 400,
  "detail": "Bad Request",
  "extra": {}
}

Query Parameters

stream_id?string|null

limit?integer

Default20

starting_after?string|null

ending_before?string|null

wait_ms?integer

Default0

Response Body

`application/json`

curl -X GET "https://example.com/v1/responses/events"

{
  "object": "list",
  "data": [
    {
      "completion_id": "string",
      "response_id": "string",
      "custom_id": "string",
      "status": "string",
      "result": {
        "property1": null,
        "property2": null
      },
      "error": {
        "property1": null,
        "property2": null
      },
      "prompt_tokens": 0,
      "completion_tokens": 0,
      "total_tokens": 0,
      "cost_micros": 0,
      "completed_at": "string"
    }
  ],
  "has_more": true
}

{
  "status_code": 400,
  "detail": "Bad Request",
  "extra": {}
}

Request Body

application/json

TypeScript Definitions

Use the request body type in TypeScript.

model*string

messages*array<>

temperature?number|null

top_p?number|null

presence_penalty?|

Range-2 <= value <= 2

frequency_penalty?|

Range-2 <= value <= 2

max_tokens?integer|null

max_completion_tokens?integer|null

top_k?|

Range-1 <= value

min_p?|

Rangevalue <= 1

repetition_penalty?|

Rangevalue <= 2

stop?string|array<string>|null

seed?integer|null

stream?boolean

Defaultfalse

stream_options?|null

tools?array<>|null

tool_choice?|string|null

response_format?|null

reasoning_effort?string|null

enable_thinking?boolean|null

prompt_cache_key?|

service_tier?|

Processing tier for the request. "flex" selects flex processing (beta) on flex-enabled accounts: discounted tokens in exchange for lower scheduling priority and best-effort completion — the request may queue significantly longer than default-tier requests, so raise your client timeout accordingly. A flex request that cannot complete in time fails with 408 (OpenAI SDKs retry it automatically). All other accepted values (auto, default, scale, priority) are served at the default tier. Responses echo the tier actually served in service_tier.

Response Body

`application/json`

curl -X POST "https://example.com/v1/chat/completions" \  -H "Content-Type: application/json" \  -d '{    "model": "string",    "messages": [      {        "role": "string"      }    ]  }'

{
  "property1": null,
  "property2": null
}

{
  "status_code": 400,
  "detail": "Bad Request",
  "extra": {}
}

Request Body

application/json

TypeScript Definitions

Use the request body type in TypeScript.

model*string

input*string|array<string>|array<integer>|array<array<integer>>

encoding_format?string

Default"float"

Value in"float" | "base64"

dimensions?integer|null

user?string|null

Response Body

`application/json`

curl -X POST "https://example.com/v1/embeddings" \  -H "Content-Type: application/json" \  -d '{    "model": "string",    "input": "string"  }'

{
  "property1": null,
  "property2": null
}

{
  "status_code": 400,
  "detail": "Bad Request",
  "extra": {}
}

Query Parameters

beta?boolean|null

Request Body

application/json

TypeScript Definitions

Use the request body type in TypeScript.

model*string

max_tokens*integer

messages*array<>

system?string|array<>|array<>|null

stop_sequences?array<string>|null

stream?boolean

Defaultfalse

temperature?number|null

top_p?number|null

top_k?integer|null

tools?array<>|null

tool_choice?|string|null

thinking?|null

enable_thinking?boolean|null

Response Body

`application/json`

curl -X POST "https://example.com/v1/messages" \  -H "Content-Type: application/json" \  -d '{    "model": "string",    "max_tokens": 0,    "messages": [      {        "role": "user",        "content": "string"      }    ]  }'

{
  "property1": null,
  "property2": null
}

{
  "status_code": 400,
  "detail": "Bad Request",
  "extra": {}
}

Response Body

`application/json`

curl -X GET "https://example.com/v1/models"

{
  "object": "list",
  "data": [
    {
      "id": "string",
      "object": "model",
      "created": 0,
      "owned_by": "sference",
      "display_name": "string",
      "modality": "text_generation",
      "capabilities": {
        "image_input": {
          "supported": false
        },
        "pdf_input": {
          "supported": false
        },
        "thinking": {
          "supported": false,
          "types": {
            "enabled": {
              "supported": false
            },
            "adaptive": {
              "supported": false
            }
          }
        },
        "tools": {
          "supported": false
        },
        "dimensions": 0
      },
      "context_tokens": 0,
      "provider": "string",
      "pricing": {
        "input_per_million_usd": 0,
        "output_per_million_usd": 0,
        "cached_input_per_million_usd": 0
      },
      "released": "string"
    }
  ]
}

API Reference

Every HTTP operation, generated from the OpenAPI specification.

Me GET

Response Body

`application/json`

Path Parameters

Response Body

`application/json`

`application/json`

Path Parameters

Response Body

`application/json`

`application/json`

Query Parameters

Response Body

`application/json`

`application/json`

Request Body

Response Body

`application/json`

`application/json`

Query Parameters

Response Body

`application/json`

`application/json`

Request Body

Response Body

`application/json`

`application/json`

Request Body

Response Body

`application/json`

`application/json`

Query Parameters

Request Body

Response Body

`application/json`

`application/json`

Response Body

`application/json`

200application/json

200application/json

400application/json

200application/json

400application/json

200application/json

400application/json

200application/json

400application/json

200application/json

400application/json

200application/json

400application/json

200application/json

400application/json

200application/json

400application/json

200application/json

`application/json`

`application/json`

`application/json`

`application/json`

`application/json`

`application/json`

`application/json`

`application/json`

`application/json`

`application/json`

`application/json`

`application/json`

`application/json`

`application/json`

`application/json`

`application/json`

`application/json`

`application/json`