> ## Documentation Index > Fetch the complete documentation index at: https://docs.powertokens.ai/llms.txt > Use this file to discover all available pages before exploring further. # deepseek-v4-flash Chat Completions > Invoke DeepSeek chat completions capability. Supported models include `deepseek-v4-flash`. Documented fields: `model`, `messages`, `thinking`, `stream`, `stream_options.include_usage`, `max_tokens`, `stop`, `reasoning_effort`, `response_format`, `temperature`, `top_p`, `logprobs`, `top_logprobs`, `tools`, `tool_choice`, `user_id`. `messages[].content` supports text content (string); `assistant` messages support `reasoning_content` and `tool_calls`. ## OpenAPI ````yaml api-reference/en/zmodelChat/deepseek/deepseek-v4-flash.json POST /v1/chat/completions openapi: 3.0.1 info: title: deepseek-v4-flash version: 1.0.0 description: >- DeepSeek Chat Completions documentation. The unified API endpoint is `/v1/chat/completions`, compatible with the OpenAI Chat Completions request format. Supported model: `deepseek-v4-flash`. This document covers the following request fields: `model`, `messages`, `messages[].content`, `messages[].tool_calls`, `messages[].reasoning_content`, `thinking`, `stream`, `stream_options.include_usage`, `max_tokens`, `stop`, `reasoning_effort`, `response_format`, `temperature`, `top_p`, `logprobs`, `top_logprobs`, `tools`, `tool_choice`, `user_id`. license: name: Project License url: https://github.com/QuantumNous/new-api/blob/main/LICENSE servers: - url: https://api.powertokens.ai description: Baze API server endpoint security: [] tags: - name: DeepSeek Chat description: DeepSeek text chat and tool calling capabilities paths: /v1/chat/completions: post: tags: - DeepSeek Chat summary: deepseek-v4-flash description: >- Invoke DeepSeek chat completions capability. Supported models include `deepseek-v4-flash`. Documented fields: `model`, `messages`, `thinking`, `stream`, `stream_options.include_usage`, `max_tokens`, `stop`, `reasoning_effort`, `response_format`, `temperature`, `top_p`, `logprobs`, `top_logprobs`, `tools`, `tool_choice`, `user_id`. `messages[].content` supports text content (string); `assistant` messages support `reasoning_content` and `tool_calls`. operationId: deepseekChatCompletions_deepseek_v4_flash requestBody: required: true content: application/json: schema: $ref: '#/components/schemas/ChatCompletionRequest' examples: default: summary: Basic text chat value: model: deepseek-v4-flash messages: - content: You are a helpful assistant role: system - content: Hi role: user thinking: type: enabled reasoning_effort: high max_tokens: 4096 response_format: type: text stop: null stream: false stream_options: null temperature: 1 top_p: 1 tools: null tool_choice: none logprobs: false top_logprobs: null responses: '200': description: >- Success. Returns JSON in non-streaming mode; returns an SSE event stream in streaming mode. content: application/json: schema: $ref: '#/components/schemas/ChatCompletionResponse' examples: success: summary: Non-streaming success response value: id: 26231-33ea-4464-af18-c8c49d256 object: chat.completion created: 1782805310 model: deepseek-v4-flash choices: - index: 0 message: role: assistant content: Hello! How can I assist you today? reasoning_content: >- We are given a conversation: "You are a helpful assistant" and "Hi". We need to respond appropriately. Since I'm the assistant, I should respond to the user's greeting. A simple friendly response is fine. logprobs: null finish_reason: stop usage: prompt_tokens: 10 completion_tokens: 55 total_tokens: 65 prompt_tokens_details: cached_tokens: 0 completion_tokens_details: reasoning_tokens: 45 prompt_cache_hit_tokens: 0 prompt_cache_miss_tokens: 10 system_fingerprint: fp_9954bca7_prod0820_fp8_kvcache_20260402 text/event-stream: schema: type: string description: >- SSE data stream. Each `data:` event can be parsed as a DeepSeek/OpenAI-compatible Chat Completions Chunk; the stream typically ends with `data: [DONE]`. When `stream_options.include_usage=true` is set, an additional chunk with `usage` is returned before the end. examples: stream: summary: Streaming response chunk value: > data: {"id":"chatcmpl_bp_123","object":"chat.completion.chunk","created":1742342400,"model":"deepseek-v4-flash","choices":[{"index":0,"delta":{"role":"assistant","content":"Calling weather tool"},"finish_reason":null}],"usage":{"prompt_tokens":48,"completion_tokens":6,"total_tokens":54}} data: {"id":"chatcmpl_bp_123","object":"chat.completion.chunk","created":1742342401,"model":"deepseek-v4-flash","choices":[{"index":0,"delta":{"tool_calls":[{"index":0,"id":"call_weather","type":"function","function":{"name":"get_weather","arguments":"{\"city\":\"Beijing\"}"}}]},"finish_reason":"tool_calls"}],"usage":{"prompt_tokens":48,"completion_tokens":22,"total_tokens":70}} data: [DONE] '400': description: Invalid request parameters. content: application/json: schema: $ref: '#/components/schemas/ErrorResponse' examples: invalid_request: summary: Parameter error value: error: message: messages is required type: invalid_request_error param: messages code: bad_request_body '401': description: Authentication failed, e.g. missing or invalid token. content: application/json: schema: $ref: '#/components/schemas/ErrorResponse' examples: unauthorized: summary: Unauthorized value: error: message: Invalid token type: api_error code: access_denied param: '' '429': description: Rate limit triggered or insufficient account quota. content: application/json: schema: $ref: '#/components/schemas/ErrorResponse' examples: quota_exceeded: summary: Insufficient quota or rate limited value: error: message: Insufficient account quota, please retry later type: api_error code: insufficient_user_quota param: '' '500': description: Internal server error while processing the request. content: application/json: schema: $ref: '#/components/schemas/ErrorResponse' examples: internal_error: summary: Internal processing failure value: error: message: Internal processing failed, please retry later type: api_error code: gen_relay_info_failed param: '' security: - BearerAuth: [] components: schemas: ChatCompletionRequest: type: object required: - model - messages description: DeepSeek chat completions request body. properties: model: type: string description: The ID of the model to use. Supports `deepseek-v4-flash`. enum: - deepseek-v4-flash example: deepseek-v4-flash messages: type: array description: A list of messages comprising the conversation so far. minItems: 1 items: $ref: '#/components/schemas/Message' thinking: $ref: '#/components/schemas/ThinkingConfig' stream: type: boolean default: false description: >- If set to true, the model will send message deltas as SSE (server-sent events). The message stream ends with `data: [DONE]`. stream_options: $ref: '#/components/schemas/StreamOptions' max_tokens: type: integer description: >- The maximum number of tokens the model can generate for the completion in a single request. The total length of input and output tokens is limited by the model's context length. stop: description: >- A string or a list of up to 16 strings. The API will stop generating further tokens when it encounters any of these strings. oneOf: - type: string - type: array items: type: string reasoning_effort: type: string enum: - high - max description: >- Controls the reasoning effort of the model. For normal requests, the default is high. For some complex agent-style requests (e.g. Claude Code, OpenCode), it is automatically set to max. For compatibility, low and medium are mapped to high, and xhigh is mapped to max. response_format: $ref: '#/components/schemas/ResponseFormat' temperature: type: number minimum: 0 maximum: 2 default: 1 description: >- Sampling temperature between 0 and 2. Higher values like 0.8 make the output more random, while lower values like 0.2 make it more focused and deterministic. We generally recommend altering this or top_p but not both. top_p: type: number minimum: 0 maximum: 1 default: 1 description: >- An alternative to sampling with temperature, the model considers the results of the tokens with top_p probability. So 0.1 means only the tokens comprising the top 10% probability mass are considered. We generally recommend altering this or temperature but not both. logprobs: type: boolean description: >- Whether to return the log probabilities of the output tokens. If true, the log probabilities of each output token are returned in the content of the message. top_logprobs: type: integer minimum: 0 maximum: 20 description: >- An integer N between 0 and 20 that specifies the top N tokens by probability to return at each output position, along with their log probabilities. When this parameter is specified, logprobs must be true. x-valid-when: logprobs: - true tools: type: array description: >- A list of tools the model may call. Currently, only function is supported as a tool. Use this parameter to provide a list of functions that take JSON as input. Up to 128 functions are supported. items: $ref: '#/components/schemas/ToolDefinition' tool_choice: $ref: '#/components/schemas/ToolChoice' user_id: type: string description: >- Your custom user_id, with the character set [a-zA-Z0-9\-_] and a maximum length of 512. Do not include private user information in user_id. user_id can be used to distinguish user identities on your side to help us with content safety review. user_id can be used for KVCache isolation for privacy management. user_id can be used for scheduling isolation of users on your business side. ChatCompletionResponse: type: object description: DeepSeek chat completions success response. properties: id: type: string description: A unique identifier for the chat. object: type: string description: The type of the object, which is chat.completion. created: type: integer description: >- The Unix timestamp (in seconds) when the chat completion was created. model: type: string description: The name of the model that generated the completion. system_fingerprint: type: string description: >- This fingerprint represents the backend configuration that the model runs with. choices: type: array description: A list of completions generated by the model. items: $ref: '#/components/schemas/ChatCompletionChoice' usage: $ref: '#/components/schemas/Usage' ErrorResponse: type: object properties: error: $ref: '#/components/schemas/ErrorObject' Message: description: >- A single message in the conversation. Depending on the role, it is divided into four message types: system, user, assistant, and tool. oneOf: - $ref: '#/components/schemas/SystemMessage' - $ref: '#/components/schemas/UserMessage' - $ref: '#/components/schemas/AssistantRequestMessage' - $ref: '#/components/schemas/ToolMessage' ThinkingConfig: type: object description: Controls the switch between thinking mode and non-thinking mode. properties: type: type: string enum: - enabled - disabled default: enabled description: >- If set to enabled, thinking mode is used. If set to disabled, non-thinking mode is used. StreamOptions: type: object description: >- Options for streaming output. This parameter can only be set when the stream parameter is true. properties: include_usage: type: boolean description: >- If set to true, an additional chunk will be transmitted just before the final `data: [DONE]` in the stream. The usage field on this chunk shows the token usage statistics for the entire request, while the choices field will always be an empty array. All other chunks will also include a usage field, but its value will be null. ResponseFormat: type: object description: >- An object specifying the format that the model must output. Setting to { "type": "json_object" } enables JSON Output, which guarantees the message the model generates is valid JSON. Important: When using JSON Output, you must also instruct the model to produce JSON yourself via a system or user message. Without this, the model may generate an unending stream of whitespace until the generation reaches the token limit, resulting in a long-running and seemingly "stuck" request. Also note that the message content may be partially cut off if finish_reason="length", which indicates the generation exceeded max_tokens or the conversation exceeded the max context length. properties: type: type: string enum: - text - json_object default: text description: Must be one of text or json_object. ToolDefinition: type: object required: - type - function description: Function tool definition. properties: type: type: string enum: - function description: The type of the tool. Currently only function is supported. function: type: object required: - name properties: name: type: string description: >- The name of the function to call. Must consist of a-z, A-Z, 0-9 characters, or include underscores and hyphens, with a maximum length of 64 characters. description: type: string description: >- A description of what the function does, used by the model to understand when and how to call it. parameters: type: object description: >- The input parameters of the function, described as a JSON Schema object. Omitting parameters defines a function with an empty parameter list. additionalProperties: true strict: type: boolean default: false description: >- If set to true, the API will use strict mode in function calls to ensure the output always conforms to the function's JSON schema definition. This is a Beta feature. ToolChoice: description: >- Controls the model's tool calling behavior. none means the model will not call any tool and instead generates a message. auto means the model can choose to generate a message or call one or more tools. required means the model must call one or more tools. Specifying a particular tool via {"type": "function", "function": {"name": "my_function"}} forces the model to call that tool. oneOf: - type: string enum: - none - auto - required - type: object required: - type - function properties: type: type: string enum: - function function: type: object required: - name properties: name: type: string description: The name of the function to call. ChatCompletionChoice: type: object properties: index: type: integer description: >- The index of this completion in the list of completions generated by the model. message: $ref: '#/components/schemas/AssistantMessage' finish_reason: type: string description: >- The reason the model stopped generating tokens. `stop` means the model finished generating naturally or hit a stop sequence; `length` means the output length reached the model's context length limit or the max_tokens limit; `content_filter` means the output was filtered due to a filtering policy; `tool_calls` means the model called a tool; `insufficient_system_resource` means generation was interrupted due to insufficient system inference resources. logprobs: $ref: '#/components/schemas/ChoiceLogprobs' Usage: type: object description: Usage information for this chat completion request. properties: prompt_tokens: type: integer description: >- The number of tokens contained in the user prompt. This value equals prompt_cache_hit_tokens + prompt_cache_miss_tokens. completion_tokens: type: integer description: The number of tokens generated by the model for the completion. total_tokens: type: integer description: The total number of tokens in this request (prompt + completion). prompt_cache_hit_tokens: type: integer description: The number of tokens in the user prompt that hit the context cache. prompt_cache_miss_tokens: type: integer description: >- The number of tokens in the user prompt that missed the context cache. completion_tokens_details: type: object description: Details about the completion tokens. properties: reasoning_tokens: type: integer description: >- The number of chain-of-thought tokens generated by the reasoning model. ErrorObject: type: object properties: message: type: string description: Error description. type: type: string description: Error type. param: type: string description: Associated field name. code: type: string description: Error code. SystemMessage: title: System message type: object required: - role - content description: System message. The role of this message is system. properties: content: type: string description: The content of the system message. role: type: string enum: - system description: The role of the message author, which is system. name: type: string description: >- An optional name for the participant, provided to the model to distinguish between participants of the same role. UserMessage: title: User message type: object required: - role - content description: User message. The role of this message is user. properties: content: type: string description: The content of the user message. role: type: string enum: - user description: The role of the message author, which is user. name: type: string description: >- An optional name for the participant, provided to the model to distinguish between participants of the same role. AssistantRequestMessage: title: Assistant message type: object required: - role description: Assistant message. The role of this message is assistant. properties: content: type: string nullable: true description: The content of the assistant message. role: type: string enum: - assistant description: The role of the message author, which is assistant. name: type: string description: >- An optional name for the participant, provided to the model to distinguish between participants of the same role. prefix: type: boolean description: >- (Beta) Set this to true to force the model to start its response with the prefix content provided in this assistant message. reasoning_content: type: string nullable: true description: >- (Beta) Used in thinking mode under conversation prefix continuation, as input for the last assistant chain-of-thought content. When using this feature, the prefix parameter must be set to true. ToolMessage: title: Tool message type: object required: - role - content - tool_call_id description: Tool message. The role of this message is tool. properties: role: type: string enum: - tool description: The role of the message author, which is tool. content: type: string description: The content of the tool message. tool_call_id: type: string description: The ID of the tool call that this message is responding to. AssistantMessage: type: object properties: role: type: string enum: - assistant description: The role that generated this message. content: type: string description: The content of this completion. reasoning_content: type: string description: >- Only applicable in thinking mode. The reasoning content in the assistant message, preceding the final answer. tool_calls: type: array description: Tool calls generated by the model, e.g. function calls. items: $ref: '#/components/schemas/ToolCallMessage' ChoiceLogprobs: type: object description: The log probabilities information for this choice. properties: content: type: array nullable: true description: A list of objects containing the log probabilities of output tokens. items: $ref: '#/components/schemas/LogprobToken' reasoning_content: type: array nullable: true description: A list of objects containing the log probabilities of output tokens. items: $ref: '#/components/schemas/LogprobToken' ToolCallMessage: type: object required: - id - type - function description: Tool call object within a message. properties: id: type: string description: The ID of the tool call. type: type: string enum: - function description: The type of the tool. Currently only function is supported. function: type: object description: The function the model called. required: - name - arguments properties: name: type: string description: The name of the function the model called. arguments: type: string description: >- The arguments to call the function with, generated by the model as a JSON string. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined in your function schema. Validate these arguments in your code before calling the function. LogprobToken: type: object description: Log probability information for an output token. properties: token: type: string description: The output token. logprob: type: number description: >- The log probability of this token. -9999.0 means the token has a very small output probability and is not among the top 20 most likely output tokens. bytes: type: array nullable: true items: type: integer description: >- A list of integers representing the UTF-8 bytes representation of the token. Useful when a UTF-8 character is split into multiple tokens. If the token has no byte representation, this value is null. top_logprobs: type: array description: >- A list of the top N tokens by output probability at this position, along with their log probabilities. In rare cases, the number of tokens returned may be fewer than the top_logprobs value specified in the request. items: $ref: '#/components/schemas/TopLogprobItem' TopLogprobItem: type: object description: >- A top log probability token for the reasoning content, containing the token, its log probability, and its UTF-8 byte representation. required: - token - logprob - bytes properties: token: type: string description: The output token. logprob: type: number description: >- The log probability of this token. -9999.0 means the token has a very small output probability and is not among the top 20 most likely output tokens. bytes: type: array nullable: true items: type: integer description: >- A list of integers representing the UTF-8 bytes representation of the token. Useful when a UTF-8 character is split into multiple tokens. If the token has no byte representation, this value is null. securitySchemes: BearerAuth: type: http scheme: bearer bearerFormat: API Token description: 'Pass `Authorization: Bearer ` in the request header.' ````