> ## Documentation Index > Fetch the complete documentation index at: https://docs.anyfast.ai/llms.txt > Use this file to discover all available pages before exploring further. # gemini-3.5-flash-lite > 使用 Gemini 3.5 Flash-Lite 生成响应。最后一个非空对话轮次必须使用 user 角色。 ## OpenAPI ````yaml zh/api-reference/model-api/google/openapi/gemini-3.5-flash-lite/openapi.yaml POST /v1beta/models/gemini-3.5-flash-lite:generateContent openapi: 3.1.0 info: title: Gemini 3.5 Flash-Lite description: 通过原生 Gemini API，根据文本和多模态输入生成文本与结构化数据。该模型已弃用 temperature、topP 和 topK。 version: 1.0.0 servers: - url: https://www.anyfast.ai security: [] paths: /v1beta/models/gemini-3.5-flash-lite:generateContent: post: summary: 生成内容 description: 使用 Gemini 3.5 Flash-Lite 生成响应。最后一个非空对话轮次必须使用 user 角色。 operationId: gemini35FlashLiteGenerateContentZh parameters: - name: key in: query required: true description: AnyFast API 密钥。 schema: type: string example: YOUR_API_KEY requestBody: required: true content: application/json: schema: type: object required: - contents properties: contents: type: array minItems: 1 description: 对话轮次。请求不能以非空 model 轮次结尾。 items: $ref: '#/components/schemas/Content' systemInstruction: $ref: '#/components/schemas/SystemInstruction' generationConfig: $ref: '#/components/schemas/GenerationConfig' tools: type: array description: 函数声明，或 Google 搜索、代码执行等受支持的 Gemini 内置工具。 items: $ref: '#/components/schemas/Tool' toolConfig: type: object additionalProperties: true description: 工具执行和函数调用配置。 safetySettings: type: array description: 按内容类别设置安全阈值。 items: $ref: '#/components/schemas/SafetySetting' example: contents: - role: user parts: - text: 将发票编号、日期和总金额提取为 JSON。 generationConfig: thinkingConfig: thinkingLevel: minimal responseMimeType: application/json maxOutputTokens: 1024 responses: '200': description: 内容生成成功 content: application/json: schema: $ref: '#/components/schemas/GenerateContentResponse' '400': description: 请求无效，包括以非空 model 轮次结尾的请求 content: application/json: schema: $ref: '#/components/schemas/Error' '401': description: 未授权 content: application/json: schema: $ref: '#/components/schemas/Error' '429': description: 请求频率超限 components: schemas: Content: type: object required: - parts properties: role: type: string enum: - user - model default: user parts: type: array minItems: 1 items: $ref: '#/components/schemas/Part' SystemInstruction: type: object properties: parts: type: array items: $ref: '#/components/schemas/Part' GenerationConfig: type: object description: 请勿传入 temperature、topP、topK、candidateCount 或 thinkingBudget。 properties: thinkingConfig: type: object properties: thinkingLevel: type: string enum: - minimal - medium - high default: minimal description: minimal 适合最高吞吐量，medium/high 适合多步工具调用。 maxOutputTokens: type: integer minimum: 1 maximum: 65536 responseMimeType: type: string example: application/json responseSchema: type: object additionalProperties: true description: 用于结构化输出的 JSON Schema。 stopSequences: type: array items: type: string Tool: type: object properties: functionDeclarations: type: array items: type: object required: - name - description - parameters properties: name: type: string description: type: string parameters: type: object additionalProperties: true description: 函数参数的 JSON Schema。 googleSearch: type: object additionalProperties: true codeExecution: type: object additionalProperties: true urlContext: type: object additionalProperties: true SafetySetting: type: object required: - category - threshold properties: category: type: string threshold: type: string GenerateContentResponse: type: object properties: candidates: type: array items: type: object properties: content: $ref: '#/components/schemas/Content' finishReason: type: string example: STOP index: type: integer example: 0 safetyRatings: type: array nullable: true items: type: object additionalProperties: true usageMetadata: type: object properties: promptTokenCount: type: integer candidatesTokenCount: type: integer cachedContentTokenCount: type: integer thoughtsTokenCount: type: integer toolUsePromptTokenCount: type: integer totalTokenCount: type: integer serviceTier: type: string example: standard promptTokensDetails: type: array items: $ref: '#/components/schemas/ModalityTokenCount' toolUsePromptTokensDetails: type: array nullable: true items: $ref: '#/components/schemas/ModalityTokenCount' modelVersion: type: string example: gemini-3.5-flash-lite responseId: type: string quota: type: integer description: AnyFast 返回的本次请求配额用量。 Error: type: object properties: error: type: object properties: code: type: integer message: type: string status: type: string Part: type: object description: 文本、媒体、函数调用或函数响应片段。 properties: text: type: string thought: type: boolean description: 该片段是否包含模型推理内容。 thoughtSignature: type: string description: 多轮工具调用中必须原样保留的不透明签名。 inlineData: type: object required: - mimeType - data properties: mimeType: type: string example: image/jpeg data: type: string description: Base64 编码的图片、音频、视频或文档数据。 fileData: type: object required: - mimeType - fileUri properties: mimeType: type: string example: application/pdf fileUri: type: string format: uri description: Gemini Files URI 或受支持的公开媒体 URL。 functionCall: type: object properties: name: type: string args: type: object additionalProperties: true functionResponse: type: object properties: name: type: string id: type: string description: 模型返回的函数调用 ID。 response: type: object additionalProperties: true executableCode: type: object properties: language: type: string example: PYTHON code: type: string description: 模型生成的可执行代码。 codeExecutionResult: type: object properties: outcome: type: string example: OUTCOME_OK output: type: string description: 代码执行输出。 ModalityTokenCount: type: object properties: modality: type: string example: TEXT tokenCount: type: integer ````