• Input Modality: Image / Text / Video

• Output Modality: Text

• Platform Capabilities: Function Calling, Structured Output, Web Search, Prefill Continuation, Context Caching, Hybrid Reasoning

I. OpenAI Compatibility — Chat Completions

Request Protocol

https

Header

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

Request URL

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

Request Body Parameters

Parameter Name	Secondary Parameters	Three-level parameter	Level 4 Parameters	Type	Required	Description
model	-	-	-	string	Required in Compatible mode	Model name. qwen3.7-plusor qwen3.7-plus-2026-05-26.
messages	-	-	-	array	Yes	Dialogue context, arranged in chronological order.
	role	-	-	string	Yes	system / user / assistant / tool.
	content	-	-	string or array	Depending on the message type	Plain text is of type string; content containing images/videos or explicit caching is of type array.
	-	type	-	string	In multi-modal/explicit caching scenarios, it is	text / image_url / video / video_url.
	-	text	-	string	When type= text, it is	Enter the text.
	-	image_url	url	string	When type= image_url, it is	Image URL or Base64 Data URL.
	-	video	-	array	When type= video, it is	Videos in the form of an image list.
	-	video_url	url	string	When type= video_url, it is	URL of the video file, or a Base64 Data URL.
	-	video_url	fps	float	No	Frames extracted per second,[0.1,10], default value: 2.0.
	-	min_pixels	-	integer	No	Minimum pixels for images/video frames. Qwen3.7 input images: default/minimum 65536;video/image list: default 65536, minimum 4096.
	-	max_pixels	-	integer	No	Image/video frame maximum pixel. Qwen3.7 input image ( vl_high_resolution_images = false): default 2621440, maximum 16777216; video/image list : default 655360, maximum 2048000.
	-	total_pixels	-	integer	No	Video frame extraction total pixel limit. Qwen 3.7 series : default/maximum 819200000.
	-	cache_control	type	string	for explicit caching	type fixed ephemeral.
	partial	-	-	boolean	No	Assistant Message. Defaults to false. Prefix continuation. qwen3.7-plus supported.
	tool_calls / tool_call_id	-	-	-	No	Fields related to Function Calling.
stream	-	-	-	boolean	No	Defaults to false. Enables streaming output.
stream_options	include_usage	-	-	boolean	No	Only valid when stream= true. Defaults to false. The usage information is returned in the last chunk.
temperature	-	-	-	float	No	[0,2). Default: non-thinking 0.7; thinking 0.6.
top_p	-	-	-	float	No	(0,1.0]. Default: non-thinking 0.8; thinking 0.95.
top_k	-	-	-	integer	No	Default 20.
repetition_penalty	-	-	-	float	No	Default 1.05.
presence_penalty	-	-	-	float	No	[-2.0,2.0]. Default: non-thinking mode 1.5; thinking mode 0.5.
response_format	type	-	-	string	Yes (if this object is passed)	text/ json_object. qwen3.7-plus supports structured output.
max_tokens	-	-	-	integer	No	Deprecated, please use max_completion_tokens instead.
max_completion_tokens	-	-	-	integer	No	Maximum tokens for complete output (chain of thought + response). The maximum output on the platform model page is 131072.
vl_high_resolution_images	-	-	-	boolean	No	Defaults to false. For the Qwen 3.7 series, when set to true, it fixes the upper limit to 16777216.
enable_thinking	-	-	-	boolean	No	Whether to enable the reasoning mode. After being enabled, it will return reasoning_content.
preserve_thinking	-	-	-	boolean	No	Defaults to false. Historical reasoning_contentis spliced back into the input. Supported by qwen3.7-plus.
thinking_budget	-	-	-	integer	No	Max tokens for reasoning. The Qwen3.7 series supports a maximum reasoning chain of 262144 tokens.
tool_stream	-	-	-	boolean	No	Defaults to false. Only valid when stream= true. qwen3.7-plus text modality supported.
enable_code_interpreter	-	-	-	boolean	No	Defaults to false. Code interpreter.
seed	-	-	-	integer	No	random seed, default 1234.
stop	-	-	-	string or array	No	Stop words.
tools / tool_choice / parallel_tool_calls	-	-	-	-	No	Function Calling related.

Request Example (Compatible · Text)

curl -X POST "https://genaiapi.cloudsway.net/v1/chat/completions" \
  -H "Authorization: Bearer ${your_ak}" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "qwen3.7-plus",
    "messages": [
      {"role": "system", "content": "You are a helpful assistant."},
      {"role": "user", "content": "Who are you?"}
    ]
  }'

Request Example (Compatible · Multimodal)

curl -X POST "https://genaiapi.cloudsway.net/v1/chat/completions" \
  -H "Authorization: Bearer ${your_ak}" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "qwen3.7-plus",
    "messages": [
      {
        "role": "user",
        "content": [
          {"type": "text", "text": "What is in this image?"},
          {"type": "image_url", "image_url": {"url": "https://example.com/image.jpg"}}
        ]
      }
    ],
    "enable_thinking": true
  }'

Request Example (Endpoint Mode)

curl -X POST "https://genaiapi.cloudsway.net/v1/ai/{endpoint}/chat/completions" \
  -H "Authorization: Bearer ${your_ak}" \
  -H "Content-Type: application/json" \
  -d '{
    "messages": [
      {"role": "user", "content": "Hello"}
    ],
    "stream": false
  }'

Example of Streaming Request

curl -X POST "https://genaiapi.cloudsway.net/v1/chat/completions" \
  -H "Authorization: Bearer ${your_ak}" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "qwen3.7-plus",
    "messages": [
      {"role": "user", "content": "Please briefly introduce artificial intelligence."}
    ],
    "stream": true,
    "stream_options": {"include_usage": true}
  }'

II. OpenAI Compatibility — Responses API

Header

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

Request URL

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

Request Body Parameters

Parameter Name	Secondary Parameters	Type	Required	Description
model	-	string	Required in Compatible mode	qwen3.7-plus/ qwen3.7-plus-2026-05-26
input	-	string or array	Yes	Plain text or message array
	role / content / type / text	-	when it is an array	EasyInputMessage: input_text / output_text
	type / image_url	-	when it is user	input_image, URL or Base64 Data URI
instructions	-	string	No	System Instructions
previous_response_id	-	string	No	Previous response ID for multi-turn dialogue (valid for 7 days)
conversation	-	string	No	Session ID, cannot be used together with previous_response_id
stream	-	boolean	No	Defaults to false
store	-	boolean	No	Defaults to true
tools	type	string	Yes (if tools are passed)	web_search / web_extractor / function etc.
tool_choice	-	string/object	No	Default: auto
temperature	-	float	No	[0, 2)
top_p	-	float	No	(0, 1.0]
enable_thinking	-	boolean	No	Mode of thought, return the reasoning output item
reasoning	effort	string	No	none/minimal/low/medium/high/xhigh/max
max_output_tokens	-	integer	No	Maximum output tokens, minimum 16

Request Example

curl -X POST "https://genaiapi.cloudsway.net/v1/responses" \
  -H "Authorization: Bearer ${your_ak}" \
  -H "Content-Type: application/json" \
  -H "x-dashscope-session-cache: enable" \
  -d '{
    "model": "qwen3.7-plus",
    "input": "What can you do?",
    "enable_thinking": true
  }'

Example of Multi-turn Dialogue

curl -X POST "https://genaiapi.cloudsway.net/v1/responses" \
  -H "Authorization: Bearer ${your_ak}" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "qwen3.7-plus",
    "input": "Explain the main idea of GBDT",
    "previous_response_id": "resp_id_returned_from_first_round"
  }'

III. Anthropic Compatibility — Messages API

Header

Parameter Name	Type	Description
Content-Type	string	application/json
Authorization	string	Bearer ${your_ak}

Request URL

https://genaiapi.cloudsway.net/v1/messages

Request Body Parameters

Parameter Name	Secondary Parameters	Three-level parameter	Type	Required	Description
model	-	-	string	Required in Compatible mode	qwen3.7-plus/ qwen3.7-plus-2026-05-26
max_tokens	-	-	integer	Yes	Max Tokens for Response, Upper Limit 131072
system	-	-	string/array	No	System prompt, which can carry cache_control
messages	role/content	-	array	Yes	user / assistant / system
	type=text / text	-	when it is an array	Text content
	type=image / source	url/base64/media_type/data	when it is an array	Image Input
	type=video / source	url/base64/media_type/data	when it is an array	Video Input
	type=tool_use / tool_result	-	It is when FC	Tool Calls and Results
stream	-	-	boolean	No	Defaults to false
temperature	-	-	number	No	[0, 2)
top_p	-	-	number	No	Nuclear Sampling
top_k	-	-	integer	No	Candidate set size
stop_sequences	-	-	array	No	stop sequence
thinking	type / budget_tokens	-	object	No	enabled/disabled; budget_tokens is to be deprecated
tools	name/description/input_schema	-	array	Yes (if tools are passed)	Function Calling
tool_choice	type / name	-	object	No	auto/any/none/tool
output_config	format.type / format.schema	-	object	No	qwen3.7 series strictly structured output (json_schema)

Request Example

curl -X POST "https://genaiapi.cloudsway.net/v1/messages" \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer ${your_ak}" \
  -d '{
    "model": "qwen3.7-plus",
    "max_tokens": 1024,
    "system": "You are a helpful assistant",
    "messages": [
      {"role": "user", "content": "Who are you?"}
    ],
    "thinking": {"type": "disabled"}
  }'

Sample Image Input

curl -X POST "https://genaiapi.cloudsway.net/v1/messages" \
  -H "Content-Type: application/json" \
  -H "Authorization: ${your_ak}" \
  -d '{
    "model": "qwen3.7-plus",
    "max_tokens": 1024,
    "messages": [{
      "role": "user",
      "content": [
        {"type": "text", "text": "Describe this image"},
        {"type": "image", "source": {"type": "url", "url": "https://example.com/a.jpg"}}
      ]
    }]
  }'