API Reference

Create Text Chat Request

POST

https://apis.iflow.cn/v1/chat/completions

Content-Type:application/json

---

Request Parameters

Authorization

• Type
  string
• Location
  header
• Required
  Yes
• Description
  Use the following format for authentication: Bearer <your api key> (visit iFlow official website to login and get API KEY).

---

LLM Models

Parameter Name	Type	Required	Default	Description
messages	object[]	Yes	-	List of messages that make up the current conversation.
messages.content	string	Yes	What opportunities and challenges will the Chinese large model industry face in 2025?	Content of the message.
messages.role	enum<string>	Yes	user	Role of the message author. Possible values: user , assistant , system
model	enum<string>	Yes	deepseek-r1	Corresponding model name. To better improve service quality, we may periodically make changes to the models provided by this service, including but not limited to model online/offline adjustments and model service capability adjustments. We will notify you through appropriate means such as announcements and message pushes when possible. Please refer to the Quick Start page for supported models.
frequency_penalty	number	No	0.5	Adjusts the frequency penalty for generated tokens to control repetitiveness.
max_tokens	integer	No	512	Maximum number of tokens to generate. Value range: 1 < x < 8192
n	integer	No	1	Number of generated results to return.
response_format	object	No	-	Object specifying the model output format.
response_format.type	string	No	-	Type of response format.
stop	string[]	null	No	-
stream	boolean	No	false	If set to true , tokens will be returned progressively as Server-Sent Events (SSE).
temperature	number	No	0.7	Controls the randomness of responses. Lower values make output more deterministic; higher values make output more random.
tools	object[]	No	-	List of tools the model may call. Currently, only functions are supported as tools. Use this parameter to provide a list of functions for which the model may generate JSON input. Supports up to 128 functions.
tools.function	object	No	-	Function object.
tools.function.name	string	No	-	Name of the function to call. Must consist of letters, numbers, underscores, or hyphens, with a maximum length of 64.
tools.function.description	string	No	-	Description of the function, used by the model to determine when and how to call it.
tools.function.parameters	object	No	-	Parameters accepted by the function, described as a JSON Schema object. If parameters are not specified, a function with an empty parameter list is defined.
tools.function.strict	boolean	null	No	false
tools.type	enum<string>	No	function	Type of tool. Currently only supports function .
top_k	number	No	50	Limits token selection range to the top k candidates.
top_p	number	No	0.7	Nucleus sampling parameter used to dynamically adjust the selection range of each predicted token based on cumulative probability.

---

Request Examples

CURL

curl --request POST \
  --url https://apis.iflow.cn/v1/chat/completions \
  --header 'Authorization: Bearer <iflow API KEY>' \
  --header 'Content-Type: application/json' \
  --data '{
  "model": "deepseek-r1",
  "messages": [
    {
      "role": "user",
      "content": "What opportunities and challenges will the Chinese large model industry face in 2025?"
    }
  ],
  "stream": false,
  "max_tokens": 512,
  "stop": [
    "null"
  ],
  "temperature": 0.7,
  "top_p": 0.7,
  "top_k": 50,
  "frequency_penalty": 0.5,
  "n": 1,
  "response_format": {
    "type": "text"
  },
  "tools": [
    {
      "type": "function",
      "function": {
        "description": "<string>",
        "name": "<string>",
        "parameters": {},
        "strict": false
      }
    }
  ]
}'

Python

import requests

url = "https://apis.iflow.cn/v1/chat/completions"

payload = {
    "model": "deepseek-r1",
    "messages": [
        {
            "role": "user",
            "content": "What opportunities and challenges will the Chinese large model industry face in 2025?"
        }
    ],
    "stream": False,
    "max_tokens": 512,
    "stop": ["null"],
    "temperature": 0.7,
    "top_p": 0.7,
    "top_k": 50,
    "frequency_penalty": 0.5,
    "n": 1,
    "response_format": {"type": "text"},
    "tools": [
        {
            "type": "function",
            "function": {
                "description": "<string>",
                "name": "<string>",
                "parameters": {},
                "strict": False
            }
        }
    ]
}
headers = {
    "Authorization": "Bearer <iFlow API KEY>",
    "Content-Type": "application/json"
}

response = requests.request("POST", url, json=payload, headers=headers)

print(response.text)

Javascript

const url = "https://apis.iflow.cn/v1/chat/completions";

const payload = {
  "model": "deepseek-r1",
  "messages": [
    {
      "role": "user",
      "content": "What opportunities and challenges will the Chinese large model industry face in 2025?"
    }
  ],
  "stream": false,
  "max_tokens": 512,
  "stop": ["null"],
  "temperature": 0.7,
  "top_p": 0.7,
  "top_k": 50,
  "frequency_penalty": 0.5,
  "n": 1,
  "response_format": {"type": "text"},
  "tools": [
    {
      "type": "function",
      "function": {
        "description": "<string>",
        "name": "<string>",
        "parameters": {},
        "strict": false
      }
    }
  ]
};

const headers = {
  "Authorization": "Bearer <iFlow API KEY>",
  "Content-Type": "application/json"
};

fetch(url, {
  method: "POST",
  headers: headers,
  body: JSON.stringify(payload)
})
.then(response => response.json())
.then(data => console.log(data));

Java

import com.fasterxml.jackson.databind.ObjectMapper;
import java.util.*;

// Build message object
Map<String, Object> message = new HashMap<>();
message.put("role", "user");
message.put("content", "What opportunities and challenges will the Chinese large model industry face in 2025?");

// Build tool function object
Map<String, Object> function = new HashMap<>();
function.put("description", "<string>");
function.put("name", "<string>");
function.put("parameters", new HashMap<>());
function.put("strict", false);

Map<String, Object> tool = new HashMap<>();
tool.put("type", "function");
tool.put("function", function);

// Build response format object
Map<String, Object> responseFormat = new HashMap<>();
responseFormat.put("type", "text");

// Build request parameters
Map<String, Object> requestBody = new HashMap<>();
requestBody.put("model", "deepseek-r1");
requestBody.put("messages", Arrays.asList(message));
requestBody.put("stream", false);
requestBody.put("max_tokens", 512);
requestBody.put("stop", Arrays.asList("null"));
requestBody.put("temperature", 0.7);
requestBody.put("top_p", 0.7);
requestBody.put("top_k", 50);
requestBody.put("frequency_penalty", 0.5);
requestBody.put("n", 1);
requestBody.put("response_format", responseFormat);
requestBody.put("tools", Arrays.asList(tool));

// Convert to JSON string
ObjectMapper mapper = new ObjectMapper();
String jsonBody = mapper.writeValueAsString(requestBody);

// Send request
HttpResponse<String> response = Unirest.post("https://apis.iflow.cn/v1/chat/completions")
  .header("Authorization", "Bearer <iFlow API KEY>")
  .header("Content-Type", "application/json")
  .body(jsonBody)
  .asString();

---

Response Parameters

Non-streaming Output

Parameter Name	Type	Required	Default	Description
choices	object[]	Yes	-	List of choices generated by the model.
choices.finish_reason	enum<string>	No	-	Reason for generation completion. Possible values: - stop : Natural completion. - eos : Reached sentence end marker. - length : Reached maximum token length limit. - tool_calls : Called a tool (such as a function).
choices.message	object	Yes	-	Message object returned by the model.
created	integer	Yes	-	Timestamp when the response was generated.
id	string	Yes	-	Unique identifier of the response.
model	string	Yes	-	Name of the model used.
object	enum<string>	Yes	-	Type of response. Possible values: - chat.completion : Indicates this is a chat completion response.
tool_calls	object[]	No	-	Tool calls generated by the model, such as function calls.
tool_calls.function	object	No	-	Function called by the model.
tool_calls.function.arguments	string	No	-	Arguments for the function call, generated by the model in JSON format. Note: The JSON generated by the model may be invalid or may generate parameters not belonging to the function definition. Please validate these parameters in your code before calling the function.
tool_calls.function.name	string	No	-	Name of the function to call.
tool_calls.id	string	No	-	Unique identifier of the tool call.
tool_calls.type	enum<string>	No	-	Type of tool. Currently only supports function . Possible values: - function : Indicates this is a function call.
usage	object	Yes	-	Token usage statistics.
usage.completion_tokens	integer	Yes	-	Number of tokens used in the completion part.
usage.prompt_tokens	integer	Yes	-	Number of tokens used in the prompt part.
usage.total_tokens	integer	Yes	-	Total number of tokens used.

Streaming Output

Parameter Name	Type	Required	Default	Description
id	string	Yes	-	Unique identifier of the chat completion. Each chunk has the same ID.
choices	object[]	Yes	-	List of choices generated by the model.
choices.finish_reason	enum<string>	No	-	Reason for generation completion. Possible values: stop (natural completion), eos (reached sentence end marker), length (reached maximum token length limit), tool_calls (called a tool, such as a function).
choices.message	object	Yes	-	Message object returned by the model.
created	integer	Yes	-	Timestamp when the response was generated (Unix timestamp in seconds).
model	string	Yes	-	Name of the model used.
object	enum<string>	Yes	-	Type of response. Possible values: chat.completion (indicating this is a chat completion response).
tool_calls	object[]	No	-	Tool calls generated by the model, such as function calls.
tool_calls.function	object	No	-	Function called by the model.
tool_calls.function.arguments	string	No	-	Arguments for the function call, generated by the model in JSON format. Note: The JSON generated by the model may be invalid or may generate parameters not belonging to the function definition. Please validate these parameters in your code before calling the function.
tool_calls.function.name	string	No	-	Name of the function to call.
tool_calls.id	string	No	-	Unique identifier of the tool call.
tool_calls.type	enum<string>	No	-	Type of tool. Currently only supports function (indicating this is a function call).
usage	object	Yes	-	Token usage statistics.
usage.completion_tokens	integer	Yes	-	Number of tokens used in the completion part.
usage.prompt_tokens	integer	Yes	-	Number of tokens used in the prompt part.
usage.total_tokens	integer	Yes	-	Total number of tokens used (prompt + completion).
delta	object	No	-	Chat completion delta generated by the streaming model response.
choices.logprobs	object or null	No	-	Log probability information for this choice.
choices.logprobs.content	array or null	No	-	List of message content tokens with log probability information.
choices.logprobs.refusal	array or null	No	-	List of refusal message tokens and their log probability information.
choices.logprobs.refusal.token	string	No	-	Token.
choices.logprobs.refusal.logprob	number	No	-	Log probability for this token if it's among the top 20 most likely tokens; otherwise uses value -9999.0 to indicate extreme unlikelihood.
choices.logprobs.refusal.bytes	array or null	No	-	List of integers representing the UTF-8 byte representation of the token. Used for cases requiring combination of multiple token byte representations.
choices.logprobs.refusal.top_logprobs	array	No	-	List of the most likely tokens and their log probabilities at the current token position. In rare cases, the returned count may be less than requested.
finish_reason	string or null	No	-	Reason why the model stopped generating tokens.
index	integer	No	-	Index of the choice in the choice list.
service_tier	string or null	No	-	Service tier used to process the request.
system_fingerprint	string	No	-	Fingerprint representing the backend configuration of the model runtime. Can be used with the seed request parameter to understand potential backend changes that might affect determinism.

---

Response Information

Non-streaming

        {
          "id": "<string>",
          "choices": [
            {
              "message": {
                "role": "assistant",
                "content": "<string>",
                "reasoning_content": "<string>"
              },
              "finish_reason": "stop"
            }
          ],
          "tool_calls": [
            {
              "id": "<string>",
              "type": "function",
              "function": {
                "name": "<string>",
                "arguments": "<string>"
              }
            }
          ],
          "usage": {
            "prompt_tokens": 123,
            "completion_tokens": 123,
            "total_tokens": 123
          },
          "created": 123,
          "model": "<string>",
          "object": "chat.completion"
        }

Streaming

    {"id":"<string>","object":"chat.completion.chunk","created":1694268190,"model":"<string>", "system_fingerprint": "fp_44709d6fcb", "choices":[{"index":0,"delta":{"role":"assistant","content":""},"logprobs":null,"finish_reason":null}]}

    {"id":"<string>","object":"chat.completion.chunk","created":1694268190,"model":"<string>", "system_fingerprint": "fp_44709d6fcb", "choices":[{"index":0,"delta":{"content":"Hello"},"logprobs":null,"finish_reason":null}]}

    ....

    {"id":"<string>","object":"chat.completion.chunk","created":1694268190,"model":"<string>", "system_fingerprint": "fp_44709d6fcb", "choices":[{"index":0,"delta":{},"logprobs":null,"finish_reason":"stop"}]}