Coverage

• Create Response（Responses API）

• Chat Completions

• Prompt Cache Session Affinity (x-grok-conv-id, Chat Only)

• Inference Parameters (reasoning/ reasoning_effort)

A. Responses / Create Response

Request Protocol

HTTPS

Header

Parameter Name	Type	Required	Description
Content-Type	string	Yes	application/json
Authorization	string	Yes	Bearer ${your_AK}

Request URL

Endpoint Interface Mode

POST https://genaiapi.cloudsway.net/v1/ai/{endpointPath}/responses

Converged Interface Mode

POST https://genaiapi.cloudsway.net/v1/responses

Request Body Parameters

Parameter Name	Secondary Parameters	Three-level parameters	Type	Required	Description
input	-	-	string | array	Yes	Input content;
model	-	-	string	Yes	Model Name;
instructions	-	-	string | null	No	System prompt; do not use together with previous_response_id
previous_response_id	-	-	string | null	No	Continue writing the response from the previous round
max_output_tokens	-	-	int | null	No	Token upper limit including output and inference
stream	-	-	bool | null	No	SSE Streaming Output
temperature	-	-	number | null	No	Sampling temperature, range 0–2; the original device may malfunction when set to 2
top_p	-	-	number | null	No	Nuclear Sampling
reasoning	-	-	object	No	Recommended writing style, including elements such as effort and summary
	effort	-	string | null	No	low / medium / high / xhigh (varies by model)
	summary	-	string | null	No	auto/ concise/ detailed (compatible field)
	generate_summary	-	string | null	No	Compatibility field, recommended to use summary
include	-	-	array | null	No	Additional output, e. g. reasoning. encrypted_content
prompt_cache_key	-	-	string | null	No	Cache Routing Key
tools	-	-	array | null	No	such as function, web search, etc. ;functionis transparently transmitted, web_search*will be stripped by the platform
	type	-	string	No	Tool type, e. g. function
	name	-	string	No	Function Name (function tool)
	description	-	string	No	Function Description Glossary:
	parameters	-	object	No	JSON Schema Parameter Definition
tool_choice	-	-	string | object	No	Tool Selection Strategy
parallel_tool_calls	-	-	bool | null	No	Whether to enable parallel tool calls
text	-	-	object	No	Output Format Configuration
	format	-	object	No	Output Format
		type	string	No	e.g. text / json_schema
		schema	object	No	JSON Schema (when type=json_schema)
user	-	-	string | null	No	End User Identifier
top_logprobs	-	-	int | null	No	0–8; grok-4.20+ may be ignored

Request Example

Non-streaming (Endpoint interface mode)

curl -X POST "https://genaiapi.cloudsway.net/v1/ai/{endpointPath}/responses" \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer ${your_AK}" \
  -d '{
    "model": "MaaS_Gr_4.6_20260812",
    "input": "Explain prompt caching in one paragraph.",
    "instructions": "You are Grok, a helpful AI assistant.",
    "max_output_tokens": 1024,
    "reasoning": {
      "effort": "medium",
      "summary": "detailed"
    },
    "store": true
  }'

Non-streaming (Integrated Interface Mode)

curl -X POST "https://genaiapi.cloudsway.net/v1/responses" \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer ${your_AK}" \
  -d '{
    "model": "MaaS_Gr_4.6_20260812",
    "input": "Explain prompt caching in one paragraph.",
    "reasoning": { "effort": "high" }
  }'

streaming

curl -N -X POST "https://genaiapi.cloudsway.net/v1/ai/{endpointPath}/responses" \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer ${your_AK}" \
  -d '{
    "model": "MaaS_Gr_4.6_20260812",
    "input": "Write a haiku about caching.",
    "stream": true,
    "reasoning": { "effort": "low" }
  }'

With Function Tool

curl -X POST "https://genaiapi.cloudsway.net/v1/ai/{endpointPath}/responses" \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer ${your_AK}" \
  -d '{
    "model": "MaaS_Gr_4.6_20260812",
    "input": "What is the weather in Tokyo?",
    "tools": [
      {
        "type": "function",
        "name": "get_weather",
        "description": "Get current weather",
        "parameters": {
          "type": "object",
          "properties": {
            "location": { "type": "string" }
          },
          "required": ["location"]
        }
      }
    ],
    "tool_choice": "auto"
  }'

Response Example

non-streaming

{
  "id": "resp_abc123xyz",
  "object": "response",
  "created_at": 1755676800,
  "model": "MaaS_Gr_4.6_20260812",
  "status": "completed",
  "output": [
    {
      "type": "message",
      "id": "msg_001",
      "role": "assistant",
      "content": [
        {
          "type": "output_text",
          "text": "Prompt caching stores KV pairs from unchanged prompt prefixes..."
        }
      ]
    }
  ],
  "usage": {
    "input_tokens": 128,
    "output_tokens": 64,
    "total_tokens": 192,
    "input_tokens_details": {
      "cached_tokens": 96
    }
  }
}

Streaming (SSE Fragment Example)

data: {"type":"response.created","response":{"id":"resp_abc123xyz","status":"in_progress"}}

data: {"type":"response.output_text.delta","delta":"Prompt caching"}

data: {"type":"response.completed","response":{"id":"resp_abc123xyz","status":"completed","usage":{"input_tokens":128,"output_tokens":64,"total_tokens":192,"input_tokens_details":{"cached_tokens":96}}}}

---

B. Chat Completions

Request Protocol

HTTPS

Header

Parameter Name	Type	Required	Description
Content-Type	string	Yes	application/json
Authorization	string	Yes	Bearer ${your_AK}
x-grok-conv-id	string	No	Prompt Cache session sticky ID; the platform extracts it from the HTTP header and transparently transmits it to the upstream (Chat Completions only)

Request URL

Endpoint Interface Mode

POST https://genaiapi.cloudsway.net/v1/ai/{endpointPath}/chat/completions

Converged Interface Mode

POST https://genaiapi.cloudsway.net/v1/chat/completions

Request Body Parameters

Parameter Name	Secondary Parameters	Three-level parameters	Level 4 Parameters	Type	Required	Description
model	-	-	-	string	Yes	Model Name;
messages	-	-	-	array	Yes	Chat Messages (Text / Image)
	role	-	-	string	Yes	system / user / assistant / tool
	content	-	-	string | array	Yes	Text or multimodal content
		type	-	string	No	Multimodal types, e. g. text/ image_url
		text	-	string	No	Text snippet
		image_url	url	string	No	Image URL
max_completion_tokens	-	-	-	int | null	No	Visible output token limit (excluding inference/tools); Grok retains this field
max_tokens	-	-	-	int | null	No	Deprecated, it is recommended to use max_completion_tokens
stream	-	-	-	bool | null	No	SSE Streaming
stream_options	-	-	-	object	No	Streaming Options
	include_usage	-	-	bool	No	Comes with usage before the end of the stream; defaults to true if not set
temperature	-	-	-	number | null	No	Sampling temperature 0–2
top_p	-	-	-	number | null	No	Nuclear Sampling
n	-	-	-	int | null	No	Number of generated items
user	-	-	-	string | null	No	End User Identifier
prompt_cache_key	-	-	-	string | null	No	Cache Sticky Routing
reasoning_effort	-	-	-	string | null	No	Inference Strength low/ medium/ high/ xhigh
tools	-	-	-	array | null	No	Function tools, maximum 128
	type	-	-	string	No	Fixed function
	function	name	-	string	No	function name
	function	description	-	string	No	Function Description
	function	parameters	-	object	No	JSON Schema
tool_choice	-	-	-	string | object	No	none / auto / required / specify function
parallel_tool_calls	-	-	-	bool | null	No	Whether to enable parallel tool calls
response_format	-	-	-	object	No	Output Format
	type	-	-	string	No	text / json_object / json_schema
	json_schema	-	-	object	No	JSON Schema Definition
frequency_penalty	-	-	-	number | null	No	-2 to 2; inference model not supported
presence_penalty	-	-	-	number | null	No	-2 to 2; grok-3 / inference model not supported
stop	-	-	-	array | null	No	Up to 4 stop strings; not supported by the inference model
deferred	-	-	-	bool | null	No	truereturns request_idfor asynchronous result retrieval (Grok extension)
reasoning_split	-	-	-	bool | null	No	Reasoning Decomposition (Grok Extension, LlmChatCompletionRequest)
tool_stream	-	-	-	bool | null	No	Tool Streaming (Grok Extension)

Request Example

Prompt Cache — Turn 1 (Cache Establishment, Endpoint Interface Mode)

curl -X POST "https://genaiapi.cloudsway.net/v1/ai/{endpointPath}/chat/completions" \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer ${your_AK}" \
  -H "x-grok-conv-id: conv_abc123" \
  -d '{
    "model": "MaaS_Gr_4.6_20260812",
    "messages": [
      {"role": "system", "content": "You are Grok, a helpful and truthful AI assistant built by xAI."},
      {"role": "user", "content": "What is prompt caching?"},
      {"role": "assistant", "content": "Prompt caching stores KV pairs from unchanged prompt prefixes so they can be reused on subsequent requests. This makes responses faster and cheaper."}
    ]
  }'

Prompt Cache — Turn 2 (Cache HIT, prefix remains unchanged, new messages appended)

curl -X POST "https://genaiapi.cloudsway.net/v1/ai/{endpointPath}/chat/completions" \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer ${your_AK}" \
  -H "x-grok-conv-id: conv_abc123" \
  -d '{
    "model": "MaaS_Gr_4.6_20260812",
    "messages": [
      {"role": "system", "content": "You are Grok, a helpful and truthful AI assistant built by xAI."},
      {"role": "user", "content": "What is prompt caching?"},
      {"role": "assistant", "content": "Prompt caching stores KV pairs from unchanged prompt prefixes so they can be reused on subsequent requests. This makes responses faster and cheaper."},
      {"role": "user", "content": "Show me a code example."}
    ]
  }'

Fusion Interface Mode + Inference + Streaming

curl -N -X POST "https://genaiapi.cloudsway.net/v1/chat/completions" \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer ${your_AK}" \
  -d '{
    "model": "MaaS_Gr_4.6_20260812",
    "messages": [
      {"role": "user", "content": "Solve: 17 * 23"}
    ],
    "reasoning_effort": "high",
    "max_completion_tokens": 2048,
    "stream": true
  }'

Response Example

non-streaming

{
  "id": "chatcmpl-abc123",
  "object": "chat.completion",
  "created": 1755676800,
  "model": "MaaS_Gr_4.6_20260812",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "Here is a simple Python example of prompt caching concepts:\n\n```python\n# Turn 1: establish prefix\nmessages = [system_msg, user_msg_1]\n# Turn 2: append new turn, prefix unchanged\nmessages.append(assistant_msg_1)\nmessages.append(user_msg_2)\n```"
      },
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 256,
    "completion_tokens": 128,
    "total_tokens": 384,
    "prompt_tokens_details": {
      "cached_tokens": 192
    }
  }
}

Streaming (SSE Fragment Example)

data: {"id":"chatcmpl-abc123","object":"chat.completion.chunk","choices":[{"index":0,"delta":{"role":"assistant","content":"Here"},"finish_reason":null}]}

data: {"id":"chatcmpl-abc123","object":"chat.completion.chunk","choices":[{"index":0,"delta":{"content":" is"},"finish_reason":null}]}

data: {"id":"chatcmpl-abc123","object":"chat.completion.chunk","choices":[{"index":0,"delta":{},"finish_reason":"stop"}],"usage":{"prompt_tokens":256,"completion_tokens":128,"total_tokens":384,"prompt_tokens_details":{"cached_tokens":192}}}

data: [DONE]