> ## Documentation Index > Fetch the complete documentation index at: https://docs.inputapi.com/llms.txt > Use this file to discover all available pages before exploring further. # Claude Messages > Create a native Claude message with content blocks and stop reasons. `POST /v1/messages` Send the gateway token through `x-api-key` and set `anthropic-version` explicitly. Native requests include `model` and `messages`; the example explicitly supplies `max_tokens`, with system instructions at top-level `system`. The schema preserves the gateway's optional budget behavior; actual native provider requests should supply a model-supported budget. Read nonstreaming `content[]` and check `stop_reason`. Values other than `end_turn` require appropriate handling: `tool_use` is not a final text answer and `max_tokens` may indicate truncation. With `stream: true`, use Claude SSE rather than a Chat / Responses event parser. [The native Claude guide and complete TypeScript example](/en/guides/native-claude) include environment variables, run commands, and failure behavior. Only common text fields are expanded here; native extensions and conversion paths depend on channel support. `/v1/messages/count_tokens` is not enabled. Verified revision: `d851a5faa49a00abaafd27bf91b311506fbfbc52`. Source files are recorded in this operation’s `x-source-files`. [The Python native SSE and tool-loop example](/en/guides/native-claude-streaming) offers text and tool modes, preserving native result association and signatures. [阅读中文](/zh/api-reference/claude-messages) ## Native cache and reasoning controls The project DTO represents request-level and content-block `cache_control`, plus top-level `thinking`, `output_config`, and `context_management`. Actual support, TTLs, budgets, and modes depend on the channel and model. Claude `context_management` is a protocol-specific object and is not copied directly during Responses conversion; identical names do not imply equivalent semantics. Native cache usage may contain `usage.cache_read_input_tokens` and `cache_creation_input_tokens`. Caching, reasoning, and compaction state do not automatically increase the model's context window. Preserve thinking / signature / redacted\_thinking continuation items according to the native protocol. Added fields verified against revision `30a6d88a4`. [Prompt caching](/en/guides/prompt-caching) · [Reasoning](/en/guides/reasoning-models) · [Context management](/en/guides/context-management) ## OpenAPI ````yaml openapi/relay.json POST /v1/messages openapi: 3.0.3 info: title: API — Core inference API version: 0.5.0 description: >- 面向接入的精选规范，不是全部管理或供应商接口。 / Curated integration specification, not the entire management or provider API. license: name: AGPL-3.0-or-later servers: - url: https://gateway.example.com description: >- Example only; replace with your gateway origin and optional path prefix, without /v1. security: - BearerAuth: [] paths: /v1/messages: post: summary: Claude 原生消息 / Native Claude message description: >- 精选文本接入字段；允许扩展字段不代表所有渠道支持。原生流式和工具扩展见指南。 / Selected text integration fields; accepting extensions does not guarantee channel support. See the guide for native streaming and tool boundaries. operationId: createClaudeMessage parameters: - name: anthropic-version in: header required: false description: 适配器缺省版本；建议显式发送 / Adapter fallback version; send explicitly. schema: type: string default: '2023-06-01' requestBody: required: true content: application/json: schema: type: object required: - model - messages properties: model: type: string minLength: 1 example: YOUR_MODEL_ID max_tokens: type: integer minimum: 0 maximum: 1073741823 example: 256 description: >- 输出预算，实际范围依赖模型；调用原生供应商时应显式设置 / Output budget; model limits apply. Set explicitly for native provider requests. system: oneOf: - type: string - type: array items: type: object required: - type properties: type: type: string text: type: string id: type: string name: type: string input: type: object additionalProperties: true tool_use_id: type: string content: oneOf: - type: string - type: array items: type: object additionalProperties: true thinking: type: string signature: type: string data: type: string is_error: type: boolean cache_control: type: object description: >- 内容块缓存标记；具体支持取决于模型与渠道 / Content-block cache marker; model and channel dependent x-source-revision: 30a6d88a4fa41009755d0f157d1656ced297803d additionalProperties: true additionalProperties: true messages: type: array minItems: 1 items: type: object required: - role - content properties: role: type: string enum: - user - assistant content: oneOf: - type: string - type: array items: type: object required: - type properties: type: type: string text: type: string id: type: string name: type: string input: type: object additionalProperties: true tool_use_id: type: string content: oneOf: - type: string - type: array items: type: object additionalProperties: true thinking: type: string signature: type: string data: type: string is_error: type: boolean cache_control: type: object description: >- 内容块缓存标记；具体支持取决于模型与渠道 / Content-block cache marker; model and channel dependent x-source-revision: 30a6d88a4fa41009755d0f157d1656ced297803d additionalProperties: true additionalProperties: true stream: type: boolean default: false tools: type: array items: type: object required: - name - input_schema properties: name: type: string description: type: string input_schema: type: object additionalProperties: true description: >- 工具参数 Schema；示例额外执行本地校验 / Tool parameter schema; examples also validate locally. additionalProperties: true cache_control: type: object description: >- 原生缓存控制；依赖渠道，TTL 与断点限制不能跨供应商假定一致 / Native cache control; channel dependent, TTL and breakpoint limits vary x-source-revision: 30a6d88a4fa41009755d0f157d1656ced297803d additionalProperties: true example: type: ephemeral thinking: type: object description: >- 原生推理配置；模式与预算由目标模型决定 / Native reasoning configuration; target model determines modes and budgets x-source-revision: 30a6d88a4fa41009755d0f157d1656ced297803d additionalProperties: true output_config: type: object description: >- 原生输出配置，包含条件支持的 effort / Native output configuration, including conditional effort support x-source-revision: 30a6d88a4fa41009755d0f157d1656ced297803d additionalProperties: true context_management: type: object description: >- Claude 专用上下文编辑对象；Claude 转 Responses 不直接复制 / Claude-specific context editing; not copied directly into Responses x-source-revision: 30a6d88a4fa41009755d0f157d1656ced297803d additionalProperties: true additionalProperties: true responses: '200': description: 检查原生完成状态及内容 / Check native completion status and content. content: application/json: schema: type: object properties: id: type: string type: type: string role: type: string model: type: string stop_reason: type: string nullable: true content: type: array items: type: object required: - type properties: type: type: string text: type: string id: type: string name: type: string input: type: object additionalProperties: true tool_use_id: type: string content: oneOf: - type: string - type: array items: type: object additionalProperties: true thinking: type: string signature: type: string data: type: string is_error: type: boolean additionalProperties: true usage: type: object additionalProperties: true properties: cache_read_input_tokens: type: integer description: 原生缓存读数；可能缺失 / Native cache reads; may be absent x-source-revision: 30a6d88a4fa41009755d0f157d1656ced297803d minimum: 0 cache_creation_input_tokens: type: integer description: 原生缓存写入读数；可能缺失 / Native cache writes; may be absent x-source-revision: 30a6d88a4fa41009755d0f157d1656ced297803d minimum: 0 additionalProperties: true example: type: message id: msg_example role: assistant stop_reason: end_turn content: - type: text text: Hello. text/event-stream: schema: type: string example: |+ event: message_stop data: {"type":"message_stop"} default: description: >- 网关或上游错误，结构可能随协议变化 / Gateway or upstream error; structure may vary by protocol. content: application/json: schema: type: object additionalProperties: true security: - ClaudeApiKey: [] - BearerAuth: [] components: securitySchemes: BearerAuth: type: http scheme: bearer description: 模型调用 API Key，不是控制台登录凭据 / Model API key, not console login credentials ClaudeApiKey: type: apiKey in: header name: x-api-key description: 网关令牌 / Gateway token ```` This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.