--- title: Create a Message url: https://platform.claude.com/docs/en/api/php/beta/messages/create --- # Create a Message `$client->beta->messages->create(int maxTokens, list messages, Model model, ?BetaCacheControlEphemeral cacheControl, ?BetaCompactionConfig compaction, ?Container container, ?BetaContextManagementConfig contextManagement, ?BetaDiagnosticsParam diagnostics, ?FallbackCreditToken fallbackCreditToken, ?BetaFallbacksParam fallbacks, ?string inferenceGeo, ?list mcpServers, ?BetaMetadata metadata, ?BetaOutputConfig outputConfig, ?BetaJSONOutputFormat outputFormat, ?ServiceTier serviceTier, ?Speed speed, ?list stopSequences, ?System system, ?BetaThinkingConfigParam thinking, ?BetaMessageThreadParam thread, ?BetaToolChoice toolChoice, ?list tools, ?list betas, ?string userProfileID, ?string workspaceID): BetaMessage` **POST** `/v1/messages` Send a structured list of input messages with text and/or image content, and the model will generate the next message in the conversation. The Messages API can be used for either single queries or stateless multi-turn conversations. Learn more about the Messages API in our [user guide](https://platform.claude.com/docs/en/get-started?sdk_languages=php) ## Parameters - `maxTokens: int` The maximum number of tokens to generate before stopping. Note that our models may stop _before_ reaching this maximum. This parameter only specifies the absolute maximum number of tokens to generate. Set to `0` to populate the [prompt cache](https://platform.claude.com/docs/en/build-with-claude/prompt-caching?sdk_languages=php#pre-warming-the-cache) without generating a response. Different models have different maximum values for this parameter. See [models](https://platform.claude.com/docs/en/about-claude/models/overview?sdk_languages=php) for details. - `messages: list` Input messages. Our models are trained to operate on alternating `user` and `assistant` conversational turns. When creating a new `Message`, you specify the prior conversational turns with the `messages` parameter, and the model then generates the next `Message` in the conversation. Consecutive `user` or `assistant` turns in your request will be combined into a single turn. Each input message must be an object with a `role` and `content`. You can specify a single `user`-role message, or you can include multiple `user` and `assistant` messages. If the final message uses the `assistant` role, the response content will continue immediately from the content in that message. This can be used to constrain part of the model's response. This is called prefill. On models that don't support prefill, creating a message that ends with a partial `assistant` response returns a 400 error. See [Prefill not supported](https://platform.claude.com/docs/en/api/errors?sdk_languages=php#prefill-not-supported). Example with a single `user` message: ```json [{"role": "user", "content": "Hello, Claude"}] ``` Example with multiple conversational turns: ```json [ {"role": "user", "content": "Hello there."}, {"role": "assistant", "content": "Hi, I'm Claude. How can I help you?"}, {"role": "user", "content": "Can you explain LLMs in plain English?"}, ] ``` Example with a partially-filled response from Claude, for models that support prefill: ```json [ {"role": "user", "content": "What's the Greek name for Sun? (A) Sol (B) Helios (C) Sun"}, {"role": "assistant", "content": "The best answer is ("}, ] ``` Each input message `content` may be either a single `string` or an array of content blocks, where each block has a specific `type`. Using a `string` for `content` is shorthand for an array of one content block of type `"text"`. The following input messages are equivalent: ```json {"role": "user", "content": "Hello, Claude"} ``` ```json {"role": "user", "content": [{"type": "text", "text": "Hello, Claude"}]} ``` See [input examples](https://platform.claude.com/docs/en/build-with-claude/working-with-messages?sdk_languages=php). Note that if you want to include a [system prompt](https://platform.claude.com/docs/en/build-with-claude/prompt-engineering/claude-prompting-best-practices?sdk_languages=php#give-claude-a-role), you can use the top-level `system` parameter. There is a limit of 100,000 messages in a single request. - `model: Model` The model that will complete your prompt. See [models](https://docs.anthropic.com/en/docs/models-overview) for additional details and options. - `cacheControl?: optional BetaCacheControlEphemeral` Top-level cache control automatically applies a cache_control marker to the last cacheable block in the request. - `compaction?: optional BetaCompactionConfig` Compaction configuration. When set on `POST /v1/messages`, the request is a compaction request: the conversation in `messages` is summarized and the response holds only the resulting `compaction` block (`stop_reason` `"compaction"`), which later requests send first in `messages` in place of the messages it summarizes. `POST /v1/messages/count_tokens` accepts this parameter and ignores it: the count it returns is for the conversation in `messages` as sent. Cannot be combined with `context_management`. - `container?: optional Container` Container identifier for reuse across requests. - `contextManagement?: optional BetaContextManagementConfig` Context management configuration. This allows you to control how Claude manages context across multiple requests, such as whether to clear function results or not. - `diagnostics?: optional BetaDiagnosticsParam` Request-level diagnostics. Supply `previous_message_id` to have the response include `diagnostics.cache_miss_reason` explaining any prompt-cache divergence from that prior request. - `fallbackCreditToken?: optional FallbackCreditToken` The `fallback_credit_token` from a prior refusal's `stop_details`. When a preceding request was refused and returned a `fallback_credit_token`, pass that code here on the retry to have the retry's cache-creation tokens for the prefix that was warm on the refused model billed at the cache-read rate. Must be redeemed by the same organization and workspace, with the same request body (optionally extended by one appended `assistant` message whose content is the partial text — with any trailing whitespace stripped from the final text block — and paired server-tool blocks streamed before the refusal; the appended-assistant form is not available for requests with `output_format` set or forced `tool_choice`), on an eligible fallback model, on the same platform, and within 5 minutes of the refusal; a mismatch is a 400. A token minted mid-server-tool-loop whose partial content was continuable may only be redeemed with the appended-assistant form — if an exact-body retry is rejected with a 400 saying the token must be redeemed by continuing the partial response, retry with the appended-assistant form instead. When the appended-assistant form is used on a model that otherwise disallows assistant-turn prefill, this token also authorizes that one prefill. - `fallbacks?: optional BetaFallbacksParam` Opt-in server-side retry on one or more substitute models when the requested model declines for policy reasons. Tried in order: if the first entry also declines, the second is tried, and so on. Some models don't support fallbacks; on those models, a list of fallback models returns a 400 error. See [Server-side fallback](https://platform.claude.com/docs/en/build-with-claude/refusals-and-fallback?sdk_languages=php#server-side-fallback). The string "default" requests the requested model's server-defined default fallback configuration. On a model that doesn't support fallbacks, the request runs on the requested model alone, so a declined request stays declined. - `inferenceGeo?: optional string` Specifies the geographic region for inference processing. If not specified, the workspace's `default_inference_geo` is used. - `mcpServers?: optional list` MCP servers to be utilized in this request - `metadata?: optional BetaMetadata` An object describing metadata about the request. - `outputConfig?: optional BetaOutputConfig` Configuration options for the model's output, such as the output format. - `serviceTier?: optional ServiceTier` Determines whether to use priority capacity (if available) or standard capacity for this request. Anthropic offers different levels of service for your API requests. See [service-tiers](https://platform.claude.com/docs/en/api/service-tiers?sdk_languages=php) for details. - `speed?: optional Speed` The inference speed mode for this request. `"fast"` enables high output-tokens-per-second inference. Some models require the `fast-mode-2026-02-01` beta header for any `speed` value. See [fast mode](https://platform.claude.com/docs/en/build-with-claude/fast-mode?sdk_languages=php) for which models. - `stopSequences?: optional list` Custom text sequences that will cause the model to stop generating. Our models will normally stop when they have naturally completed their turn, which will result in a response `stop_reason` of `"end_turn"`. If you want the model to stop generating when it encounters custom strings of text, you can use the `stop_sequences` parameter. If the model encounters one of the custom sequences, the response `stop_reason` value will be `"stop_sequence"` and the response `stop_sequence` value will contain the matched stop sequence. - `system?: optional System` System prompt. A system prompt is a way of providing context and instructions to Claude, such as specifying a particular goal or role. See our [guide to system prompts](https://platform.claude.com/docs/en/build-with-claude/prompt-engineering/claude-prompting-best-practices?sdk_languages=php#give-claude-a-role). - `thinking?: optional BetaThinkingConfigParam` Configuration for Claude's thinking. With `{"type": "adaptive"}`, Claude decides when and how much to think. With `{"type": "enabled"}` (manual extended thinking), you set a `budget_tokens` of at least 1,024. Thinking tokens count toward your `max_tokens` limit. Which `type` values are accepted, and what happens when you omit `thinking`, depend on the model. See [thinking](https://platform.claude.com/docs/en/build-with-claude/thinking?sdk_languages=php#configuring-thinking) for each model's behavior. - `thread?: optional BetaMessageThreadParam` Server-held conversation state (a Thread). `{"type": "create"}` asks the server to hold this conversation's state under the response's `id`; `{"type": "continue", "previous_message_id": ...}` resumes from a previous response so only the new turn's messages are sent. The response shape is unchanged. - `toolChoice?: optional BetaToolChoice` How the model should use the provided tools. The model can use a specific tool, any available tool, decide by itself, or not use tools at all. - `tools?: optional list` Definitions of tools that the model may use. If you include `tools` in your API request, the model may return `tool_use` content blocks that represent the model's use of those tools. You can then run those tools using the tool input generated by the model and then optionally return results back to the model using `tool_result` content blocks. There are two types of tools: **client tools** and **server tools**. The behavior described below applies to client tools. For [server tools](https://platform.claude.com/docs/en/agents-and-tools/tool-use/server-tools?sdk_languages=php), see their individual documentation as each has its own behavior (e.g., the [web search tool](https://platform.claude.com/docs/en/agents-and-tools/tool-use/web-search-tool?sdk_languages=php)). Each tool definition includes: * `name`: Name of the tool. * `description`: Optional, but strongly-recommended description of the tool. * `input_schema`: [JSON schema](https://json-schema.org/draft/2020-12) for the tool `input` shape that the model will produce in `tool_use` output content blocks. For example, if you defined `tools` as: ```json [ { "name": "get_stock_price", "description": "Get the current stock price for a given ticker symbol.", "input_schema": { "type": "object", "properties": { "ticker": { "type": "string", "description": "The stock ticker symbol, e.g. AAPL for Apple Inc." } }, "required": ["ticker"] } } ] ``` And then asked the model "What's the S&P 500 at today?", the model might produce `tool_use` content blocks in the response like this: ```json [ { "type": "tool_use", "id": "toolu_01D7FLrfh4GYq7yT1ULFeyMV", "name": "get_stock_price", "input": { "ticker": "^GSPC" } } ] ``` You might then run your `get_stock_price` tool with `{"ticker": "^GSPC"}` as an input, and return the following back to the model in a subsequent `user` message: ```json [ { "type": "tool_result", "tool_use_id": "toolu_01D7FLrfh4GYq7yT1ULFeyMV", "content": "259.75 USD" } ] ``` Tools can be used for workflows that include running client-side tools and functions, or more generally whenever you want the model to produce a particular JSON structure of output. See our [guide](https://platform.claude.com/docs/en/agents-and-tools/tool-use/overview?sdk_languages=php) for more details. - `betas?: optional list` (header parameter) Optional header to specify the beta version(s) you want to use. - `userProfileID?: optional string` (header parameter) The user profile ID to attribute this request to. Use when acting on behalf of a party other than your organization. - `workspaceID?: optional string` (header parameter) Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`). Only needed for credentials that can act on more than one Workspace. A credential that belongs to a specific Workspace may omit it; if sent, it must match that Workspace. - `outputFormat?: optional BetaJSONOutputFormat` **Deprecated** Deprecated: Use `output_config.format` instead. See [structured outputs](https://platform.claude.com/docs/en/build-with-claude/structured-outputs?sdk_languages=php) A schema to specify Claude's output format in responses. This parameter will be removed in a future release. ## Returns - `class BetaMessage` - `"message" type` Object type. For Messages, this is always `"message"`. - `string id` Unique object identifier. The format and length of IDs may change over time. - `?BetaContainer container` Information about the container used in this request. This will be non-null if a container tool (e.g. code execution) was used. - `list content` Content generated by the model. This is an array of content blocks, each of which has a `type` that determines its shape. Example: ```json [{"type": "text", "text": "Hi, I'm Claude."}] ``` If the request input `messages` ended with an `assistant` turn, then the response `content` will continue directly from that last turn. You can use this to constrain the model's output. For example, if the input `messages` were: ```json [ {"role": "user", "content": "What's the Greek name for Sun? (A) Sol (B) Helios (C) Sun"}, {"role": "assistant", "content": "The best answer is ("} ] ``` Then the response `content` might be: ```json [{"type": "text", "text": "B)"}] ``` - `?BetaContextManagementResponse contextManagement` Context management response. Information about context management strategies applied during the request. - `?BetaDiagnostics diagnostics` Request-level diagnostics. `null` when the request did not supply `diagnostics`, or when it did and no prompt-cache divergence was detected. - `Model model` The model that will complete your prompt. See [models](https://docs.anthropic.com/en/docs/models-overview) for additional details and options. - `"assistant" role` Conversational role of the generated message. This will always be `"assistant"`. - `?BetaRefusalStopDetails stopDetails` Structured information about why model output stopped. This is `null` when the `stop_reason` has no additional detail to report. - `?BetaStopReason stopReason` The reason that we stopped. This may be one the following values: * `"end_turn"`: the model reached a natural stopping point * `"max_tokens"`: we exceeded the requested `max_tokens` or the model's maximum * `"stop_sequence"`: one of your provided custom `stop_sequences` was generated * `"tool_use"`: the model invoked one or more tools * `"pause_turn"`: we paused a long-running turn. You may provide the response back as-is in a subsequent request to let the model continue. * `"refusal"`: when streaming classifiers intervene to handle potential policy violations * `"model_context_window_exceeded"`: we exceeded the model's context window In non-streaming mode this value is always non-null. In streaming mode, it is null in the `message_start` event and non-null otherwise. - `?string stopSequence` Which custom stop sequence was generated, if any. This value will be a non-null string if one of your custom stop sequences was generated. - `BetaUsage usage` Billing and rate-limit usage. Anthropic's API bills and rate-limits by token counts, as tokens represent the underlying cost to our systems. Under the hood, the API transforms requests into a format suitable for the model. The model's output then goes through a parsing stage before becoming an API response. As a result, the token counts in `usage` will not match one-to-one with the exact visible content of an API request or response. For example, `output_tokens` will be non-zero, even for an empty string response from Claude. Total input tokens in a request is the summation of `input_tokens`, `cache_creation_input_tokens`, and `cache_read_input_tokens`. - `?list inputTransformations` Changes the API made to the request's input before showing it to the model, and blocks that failed a binding check but were left unchanged: one entry per block, in request order. Two entry types today. `thinking_dropped` — a `thinking`, `redacted_thinking` or `connector_text` block from the request's `messages` that was removed from the prompt instead of being shown to the model because it failed a binding check. `thinking_mismatch_allowed` — a `thinking` or `redacted_thinking` block that failed the conversation check (the conversation before it differs from the one it was created in, or it carries no record of one on a model that requires it) and was shown to the model all the same, because that check is not enforced for this request. More entry types may be added over time; ignore types you do not recognize. Requires `anthropic-beta: thinking-binding-controls-2026-08-01`. Present on every such response from a model that supports extended thinking, as `[]` when there is no entry to report; without the beta, blocks are removed or left in place all the same but nothing is reported. Removed blocks contribute nothing to `usage.input_tokens`; blocks left in place count as sent. When streaming, the array is final in `message_start`; the final `message_delta` event carries it only when a server-side model fallback happened mid-stream, in which case it holds the serving model's entries and replaces the one in `message_start`. - `class BetaRawMessageStreamEvent` - `class BetaRawMessageStartEvent` - `"message_start" type` - `BetaMessage message` - `class BetaRawMessageDeltaEvent` - `"message_delta" type` - `?BetaContextManagementResponse contextManagement` Information about context management strategies applied during the request - `Delta delta` - `BetaMessageDeltaUsage usage` Billing and rate-limit usage. Anthropic's API bills and rate-limits by token counts, as tokens represent the underlying cost to our systems. Under the hood, the API transforms requests into a format suitable for the model. The model's output then goes through a parsing stage before becoming an API response. As a result, the token counts in `usage` will not match one-to-one with the exact visible content of an API request or response. For example, `output_tokens` will be non-zero, even for an empty string response from Claude. Total input tokens in a request is the summation of `input_tokens`, `cache_creation_input_tokens`, and `cache_read_input_tokens`. - `?list inputTransformations` Changes the API made to the request's input before showing it to the model, and blocks that failed a binding check but were left unchanged: one entry per block, in request order. Two entry types today. `thinking_dropped` — a `thinking`, `redacted_thinking` or `connector_text` block from the request's `messages` that was removed from the prompt instead of being shown to the model because it failed a binding check. `thinking_mismatch_allowed` — a `thinking` or `redacted_thinking` block that failed the conversation check (the conversation before it differs from the one it was created in, or it carries no record of one on a model that requires it) and was shown to the model all the same, because that check is not enforced for this request. More entry types may be added over time; ignore types you do not recognize. Requires `anthropic-beta: thinking-binding-controls-2026-08-01`. Present on every such response from a model that supports extended thinking, as `[]` when there is no entry to report; without the beta, blocks are removed or left in place all the same but nothing is reported. Removed blocks contribute nothing to `usage.input_tokens`; blocks left in place count as sent. When streaming, the array is final in `message_start`; the final `message_delta` event carries it only when a server-side model fallback happened mid-stream, in which case it holds the serving model's entries and replaces the one in `message_start`. - `class BetaRawMessageStopEvent` - `"message_stop" type` - `class BetaRawContentBlockStartEvent` - `"content_block_start" type` - `ContentBlock contentBlock` - `int index` - `class BetaRawContentBlockDeltaEvent` - `"content_block_delta" type` - `BetaRawContentBlockDelta delta` - `int index` - `class BetaRawContentBlockStopEvent` - `"content_block_stop" type` - `int index` ## Example ```php beta->messages->create( maxTokens: 1024, messages: [['content' => 'Hello, world', 'role' => 'user']], model: Model::CLAUDE_OPUS_5_5, cacheControl: ['type' => 'ephemeral', 'ttl' => '5m'], compaction: ['type' => 'summarize', 'instructions' => 'instructions'], container: [ 'id' => 'id', 'skills' => [ ['skillID' => 'pdf', 'type' => 'anthropic', 'version' => 'latest'] ], ], contextManagement: [ 'edits' => [ [ 'type' => 'clear_tool_uses_20250919', 'clearAtLeast' => ['type' => 'input_tokens', 'value' => 0], 'clearToolInputs' => true, 'excludeTools' => ['string'], 'keep' => ['type' => 'tool_uses', 'value' => 0], 'trigger' => ['type' => 'input_tokens', 'value' => 1], ], ], ], diagnostics: ['previousMessageID' => 'previous_message_id'], fallbackCreditToken: 'x', fallbacks: 'default', inferenceGeo: 'inference_geo', mcpServers: [ [ 'name' => 'name', 'type' => 'url', 'url' => 'url', 'authorizationToken' => 'authorization_token', 'toolConfiguration' => ['allowedTools' => ['string'], 'enabled' => true], ], ], metadata: ['userID' => '13803d75-b4b5-4c3e-b2a2-6f21399b021b'], outputConfig: [ 'effort' => 'low', 'format' => ['schema' => ['foo' => 'bar'], 'type' => 'json_schema'], 'taskBudget' => ['total' => 1024, 'type' => 'tokens', 'remaining' => 0], ], outputFormat: ['schema' => ['foo' => 'bar'], 'type' => 'json_schema'], serviceTier: 'auto', speed: 'standard', stopSequences: ['string'], system: [ [ 'text' => 'Today\'s date is 2024-06-01.', 'type' => 'text', 'cacheControl' => ['type' => 'ephemeral', 'ttl' => '5m'], 'citations' => [ [ 'citedText' => 'The grass is green. The sky is blue.', 'documentIndex' => 0, 'documentTitle' => 'x', 'endCharIndex' => 0, 'startCharIndex' => 0, 'type' => 'char_location', ], ], ], ], thinking: [ 'type' => 'adaptive', 'blockBinding' => [ 'prefixMismatchBehavior' => BetaThinkingPrefixMismatchBehavior::ERROR ], 'display' => 'summarized', ], thread: ['type' => 'create'], toolChoice: ['type' => 'auto', 'disableParallelToolUse' => true], tools: [ [ 'inputSchema' => [ 'type' => 'object', 'properties' => ['location' => 'bar', 'unit' => 'bar'], 'required' => ['location'], ], 'name' => 'name', 'allowedCallers' => ['direct'], 'cacheControl' => ['type' => 'ephemeral', 'ttl' => '5m'], 'deferLoading' => true, 'description' => 'Get the current weather in a given location', 'eagerInputStreaming' => true, 'inputExamples' => [['foo' => 'bar']], 'strict' => true, 'type' => 'custom', ], ], betas: [AnthropicBeta::MESSAGE_BATCHES_2024_09_24], userProfileID: 'anthropic-user-profile-id', workspaceID: 'wrkspc_011CZkZaBF1tNoB5wlCeusgy', ); var_dump($betaMessage); ``` ### Response (200) ```json { "id": "msg_013Zva2CMHLNnXjNJJKqJ2EF", "container": { "id": "container_011CpZohnwH4vuy7gazohgSP", "expires_at": "2019-12-27T18:11:19.117Z", "skills": [ { "skill_id": "pdf", "type": "anthropic", "version": "latest" } ] }, "content": [ { "citations": [ { "cited_text": "The grass is green. The sky is blue.", "document_index": 0, "document_title": "My Document", "end_char_index": 0, "file_id": "file_011CNha8iCJcU1wXNR6q4V8w", "start_char_index": 0, "type": "char_location" } ], "text": "Hi! My name is Claude.", "type": "text" } ], "context_management": { "applied_edits": [ { "cleared_input_tokens": 0, "cleared_tool_uses": 0, "type": "clear_tool_uses_20250919" } ] }, "diagnostics": { "cache_miss_reason": { "cache_missed_input_tokens": 0, "type": "model_changed" } }, "model": "claude-opus-5-5", "role": "assistant", "stop_details": { "category": "cyber", "explanation": "This request was declined because it conflicts with Anthropic's Usage Policy.", "fallback_credit_token": "QW50aHJvcGljL0NsYXVkZQ==", "fallback_has_prefill_claim": true, "recommended_model": "claude-opus-4-8", "type": "refusal" }, "stop_reason": "end_turn", "stop_sequence": null, "type": "message", "usage": { "cache_creation": { "ephemeral_1h_input_tokens": 0, "ephemeral_5m_input_tokens": 0 }, "cache_creation_input_tokens": 2051, "cache_read_input_tokens": 2051, "fallback_credit": { "status": { "type": "redeemed" } }, "inference_geo": "global", "input_tokens": 2095, "iterations": [ { "cache_creation": { "ephemeral_1h_input_tokens": 0, "ephemeral_5m_input_tokens": 0 }, "cache_creation_input_tokens": 0, "cache_read_input_tokens": 0, "input_tokens": 0, "model": "claude-haiku-5-5", "output_tokens": 0, "type": "message" } ], "output_tokens": 503, "output_tokens_details": { "thinking_tokens": 0 }, "server_tool_use": { "web_fetch_requests": 2, "web_search_requests": 0 }, "service_tier": "standard", "speed": "standard" }, "input_transformations": [ { "path": "path", "reason": "model_binding_mismatch", "type": "thinking_dropped" } ] } ```