> ## Documentation Index
> Fetch the complete documentation index at: https://docs.pinkfish.ai/llms.txt
> Use this file to discover all available pages before exploring further.
# deepseek
> DeepSeek AI API
**Server path:** `/deepseek` | **Type:** Application | **PCID required:** Yes
## Tools
| Tool | Description |
| --------------------------------------------------------------------- | ------------------------------ |
| [`deepseek_create_chat_completion`](#deepseek_create_chat_completion) | Create a chat completion |
| [`deepseek_create_completion`](#deepseek_create_completion) | Create a text completion (FIM) |
| [`deepseek_get_user_balance`](#deepseek_get_user_balance) | Get user balance |
| [`deepseek_list_models`](#deepseek_list_models) | List available models |
***
## deepseek\_create\_chat\_completion
Create a chat completion
**Parameters:**
| Parameter | Type | Required | Default | Description | |
| ------------------ | --------- | -------- | ------- | ------------------------------------------------------------------------------------------------------------------------------------ | ------------------------------------------------------------------------------------------------------ |
| `logprobs` | boolean | null | No | — | Whether to return log probabilities of the output tokens. |
| `max_tokens` | integer | null | No | — | The maximum number of tokens to generate in the response. |
| `messages` | object\[] | Yes | — | A list of messages comprising the conversation so far. | |
| `model` | string | Yes | — | ID of the model to use. | |
| `reasoning_effort` | string | null | No | — | Controls how much reasoning effort the model uses. Top-level parameter alongside thinking. |
| `response_format` | object | No | — | An object specifying the format of the response. | |
| `stop` | string\[] | No | — | Up to 16 sequences where the API will stop generating further tokens. | |
| `stream` | boolean | null | No | — | If true, partial message deltas will be sent as server-sent events. |
| `stream_options` | object | No | — | Options for streaming responses. | |
| `temperature` | number | null | No | — | Sampling temperature between 0 and 2. Higher values make output more random. |
| `thinking` | object | No | — | Configuration for the thinking/reasoning mode. | |
| `tool_choice` | object | No | — | Controls which tool is called by the model. 'none' disables tool calls, 'auto' lets the model decide, 'required' forces a tool call. | |
| `tools` | object\[] | No | — | A list of tools the model may call. Up to 128 functions. | |
| `top_logprobs` | integer | null | No | — | Number of most likely tokens to return at each position (0-20). Requires logprobs to be true. |
| `top_p` | number | null | No | — | Nucleus sampling parameter. The model considers tokens with top\_p probability mass. |
| `user` | string | null | No | — | A unique identifier representing your end-user. Max 512 characters, alphanumeric, dash, or underscore. |
```json theme={null}
{
"type": "object",
"properties": {
"PCID": {
"type": "string",
"description": "Pink Connect ID for the authenticated connection"
},
"logprobs": {
"type": [
"boolean",
"null"
],
"description": "Whether to return log probabilities of the output tokens."
},
"max_tokens": {
"type": [
"integer",
"null"
],
"description": "The maximum number of tokens to generate in the response."
},
"messages": {
"type": "array",
"items": {
"type": "object",
"properties": {
"role": {
"type": "string",
"enum": [
"system",
"user",
"assistant",
"tool"
],
"description": "The role of the message author."
},
"content": {
"type": "string",
"description": "The content of the message."
},
"name": {
"type": "string",
"description": "An optional name for the participant."
},
"tool_call_id": {
"type": "string",
"description": "Tool call that this message is responding to. Required for tool messages."
},
"tool_calls": {
"type": "array",
"items": {
"type": "object",
"properties": {
"id": {
"type": "string",
"description": "The ID of the tool call."
},
"type": {
"type": "string"
},
"function": {
"type": "object"
}
}
},
"description": "Tool calls generated by the model."
},
"prefix": {
"type": "boolean",
"description": "Beta feature. Set true on the last assistant message to force the model to continue from that prefix."
},
"reasoning_content": {
"type": "string",
"description": "The reasoning content from thinking mode. Used when passing back reasoning in multi-turn."
}
},
"required": [
"role"
]
},
"description": "A list of messages comprising the conversation so far."
},
"model": {
"type": "string",
"description": "ID of the model to use.",
"enum": [
"deepseek-v4-flash",
"deepseek-v4-pro"
]
},
"reasoning_effort": {
"type": [
"string",
"null"
],
"description": "Controls how much reasoning effort the model uses. Top-level parameter alongside thinking.",
"enum": [
"high",
"max"
]
},
"response_format": {
"type": "object",
"description": "An object specifying the format of the response.",
"properties": {
"type": {
"type": "string",
"description": "The format type.",
"enum": [
"text",
"json_object"
]
}
}
},
"stop": {
"type": "array",
"items": {
"type": "string"
},
"description": "Up to 16 sequences where the API will stop generating further tokens."
},
"stream": {
"type": [
"boolean",
"null"
],
"description": "If true, partial message deltas will be sent as server-sent events."
},
"stream_options": {
"type": "object",
"description": "Options for streaming responses.",
"properties": {
"include_usage": {
"type": "boolean",
"description": "If true, an additional chunk with token usage statistics is sent before the [DONE] message."
}
}
},
"temperature": {
"type": [
"number",
"null"
],
"description": "Sampling temperature between 0 and 2. Higher values make output more random."
},
"thinking": {
"type": "object",
"description": "Configuration for the thinking/reasoning mode.",
"properties": {
"type": {
"type": "string",
"description": "Whether thinking is enabled or disabled.",
"enum": [
"enabled",
"disabled"
]
}
}
},
"tool_choice": {
"description": "Controls which tool is called by the model. 'none' disables tool calls, 'auto' lets the model decide, 'required' forces a tool call."
},
"tools": {
"type": "array",
"items": {
"type": "object",
"properties": {
"type": {
"type": "string",
"enum": [
"function"
],
"description": "The type value"
},
"function": {
"type": "object",
"description": "The function value"
}
},
"required": [
"type",
"function"
]
},
"description": "A list of tools the model may call. Up to 128 functions."
},
"top_logprobs": {
"type": [
"integer",
"null"
],
"description": "Number of most likely tokens to return at each position (0-20). Requires logprobs to be true."
},
"top_p": {
"type": [
"number",
"null"
],
"description": "Nucleus sampling parameter. The model considers tokens with top_p probability mass."
},
"user": {
"type": [
"string",
"null"
],
"description": "A unique identifier representing your end-user. Max 512 characters, alphanumeric, dash, or underscore."
}
},
"required": [
"PCID",
"messages",
"model"
]
}
```
***
## deepseek\_create\_completion
Create a text completion (FIM)
**Parameters:**
| Parameter | Type | Required | Default | Description | |
| ---------------- | --------- | -------- | ------- | ---------------------------------------------------------------------- | ------------------------------------------------------------------------ |
| `echo` | boolean | No | — | If true, echo back the prompt in addition to the completion. | |
| `logprobs` | integer | null | No | — | Include the log probabilities on the most likely output tokens (0-20). |
| `max_tokens` | integer | null | No | — | The maximum number of tokens to generate. |
| `model` | string | Yes | — | ID of the model to use. Only deepseek-v4-pro supports FIM completions. | |
| `prompt` | string | Yes | — | The prompt to generate completions for. | |
| `stop` | string\[] | No | — | Up to 16 sequences where the API will stop generating. | |
| `stream` | boolean | No | — | If true, partial completion tokens will be sent as server-sent events. | |
| `stream_options` | object | No | — | Options for streaming responses. | |
| `suffix` | string | null | No | — | The suffix that comes after the completion. Used for fill-in-the-middle. |
| `temperature` | number | No | — | Sampling temperature between 0 and 2. | |
| `top_p` | number | No | — | Nucleus sampling parameter. | |
```json theme={null}
{
"type": "object",
"properties": {
"PCID": {
"type": "string",
"description": "Pink Connect ID for the authenticated connection"
},
"echo": {
"type": "boolean",
"description": "If true, echo back the prompt in addition to the completion."
},
"logprobs": {
"type": [
"integer",
"null"
],
"description": "Include the log probabilities on the most likely output tokens (0-20)."
},
"max_tokens": {
"type": [
"integer",
"null"
],
"description": "The maximum number of tokens to generate."
},
"model": {
"type": "string",
"description": "ID of the model to use. Only deepseek-v4-pro supports FIM completions.",
"enum": [
"deepseek-v4-pro"
]
},
"prompt": {
"type": "string",
"description": "The prompt to generate completions for."
},
"stop": {
"type": "array",
"items": {
"type": "string"
},
"description": "Up to 16 sequences where the API will stop generating."
},
"stream": {
"type": "boolean",
"description": "If true, partial completion tokens will be sent as server-sent events."
},
"stream_options": {
"type": "object",
"description": "Options for streaming responses.",
"properties": {
"include_usage": {
"type": "boolean",
"description": "If true, include usage statistics before the [DONE] message."
}
}
},
"suffix": {
"type": [
"string",
"null"
],
"description": "The suffix that comes after the completion. Used for fill-in-the-middle."
},
"temperature": {
"type": "number",
"description": "Sampling temperature between 0 and 2."
},
"top_p": {
"type": "number",
"description": "Nucleus sampling parameter."
}
},
"required": [
"PCID",
"model",
"prompt"
]
}
```
***
## deepseek\_get\_user\_balance
Get user balance
```json theme={null}
{
"type": "object",
"properties": {
"PCID": {
"type": "string",
"description": "Pink Connect ID for the authenticated connection"
}
},
"required": [
"PCID"
]
}
```
***
## deepseek\_list\_models
List available models
```json theme={null}
{
"type": "object",
"properties": {
"PCID": {
"type": "string",
"description": "Pink Connect ID for the authenticated connection"
}
},
"required": [
"PCID"
]
}
```