SPB Git

Repository Public

errors.md 27.9 KB 539 lines markdown

1---2title: Claude API errors3url: https://platform.claude.com/docs/en/api/errors4description: Understand the HTTP status codes, error response shape, and request IDs the Claude API returns, and handle errors with the SDKs' typed exceptions.5---67## HTTP errors89The API follows a predictable HTTP error code format:1011* 400 - `invalid_request_error`: There was an issue with the format or content of your request. This error type may also be used for other 4XX status codes not listed in this section. The API also returns a 400 when usage reaches an organization or workspace [spend limit you set](https://platform.claude.com/docs/en/api/rate-limits#setting-your-own-spend-limit), except limits on the [Claude Code workspace](https://platform.claude.com/docs/en/manage-claude/workspaces#claude-code-workspace), which can return a 429 instead.1213* 401 - `authentication_error`: There's an issue with your [API key](https://platform.claude.com/docs/en/get-api-key) (for example, it's malformed, revoked, or expired; see [Key expiration](https://platform.claude.com/docs/en/manage-claude/authentication#key-expiration)). On Claude Platform on AWS, this can also indicate a problem with your AWS credentials or SigV4 signature.1415* 402 - `billing_error`: There's an issue with your billing or payment information. Check your payment details in the [Claude Console](https://platform.claude.com), or in AWS Marketplace if you're using Claude Platform on AWS.1617* 403 - `permission_error`: Your API key does not have permission to use the specified resource. Check your organization's access and workspace settings in the [Claude Console](https://platform.claude.com).1819* 404 - `not_found_error`: The requested resource was not found. Check the endpoint path and any resource IDs in the request URL.2021* 409 - `conflict_error`: The request conflicts with the current state of a resource. For example, the resource was modified concurrently, or a value that must be unique is already in use. Resolve the conflict, then retry the request.2223* 413 - `request_too_large`: Request exceeds the maximum allowed number of bytes. See [Request size limits](https://platform.claude.com/docs/en/api/errors#request-size-limits) for per-endpoint maximums.2425* 429 - `rate_limit_error`: Your organization has hit a [rate limit](https://platform.claude.com/docs/en/api/rate-limits), reached its usage tier's monthly spend cap, or reached a spend limit on the Claude Code workspace. A tier spend-cap 429 has no `retry-after` header and keeps failing until access resumes; see [Reaching your spend cap](https://platform.claude.com/docs/en/api/rate-limits#reaching-your-spend-cap) for how to recognize it.2627* 500 - `api_error`: An unexpected error has occurred internal to Anthropic's systems. Retry the request with exponential backoff; if the error persists, contact support with the [request ID](https://platform.claude.com/docs/en/api/errors#request-id).2829* 504 - `timeout_error`: The request timed out while processing. Consider using the [streaming Messages API](https://platform.claude.com/docs/en/build-with-claude/streaming) for long-running requests. See [Long requests](https://platform.claude.com/docs/en/api/errors#long-requests) for more options.3031* 529 - `overloaded_error`: The API is temporarily overloaded.3233  <Warning>34    529 errors can occur when the API experiences high traffic across all users.3536    In rare cases, if your organization has a sharp increase in usage, you might see 429 errors because of acceleration limits on the API. To avoid hitting acceleration limits, ramp up your traffic gradually and maintain consistent usage patterns.37  </Warning>3839The official SDKs automatically retry transient failures (such as connection errors, rate limits, and 5xx server errors) with exponential backoff, twice by default, honoring the `retry-after` header when present. Each SDK client accepts a maximum-retries option to configure or disable this behavior.4041When receiving a [streaming](https://platform.claude.com/docs/en/build-with-claude/streaming) response over server-sent events (SSE), an error can occur after the API returns a 200 response. In that case, error handling doesn't follow these standard mechanisms. See [Error events](https://platform.claude.com/docs/en/build-with-claude/streaming#error-events) for the shape of mid-stream errors.4243## Request size limits4445The API enforces request size limits:4647| Endpoint type                                                                       | Maximum request size |48| ----------------------------------------------------------------------------------- | -------------------- |49| Messages API                                                                        | 32 MB                |50| Token Counting API                                                                  | 32 MB                |51| [Batch API](https://platform.claude.com/docs/en/build-with-claude/batch-processing) | 256 MB               |52| [Files API](https://platform.claude.com/docs/en/build-with-claude/files)            | 500 MB               |5354If you exceed these limits, you'll receive a 413 `request_too_large` error. On the direct Claude API, Cloudflare returns this error before the request reaches the API servers.5556## Error shapes5758The API always returns errors as JSON, with a top-level `error` object that always includes a `type` and `message` value. The response also includes a `request_id` field for easier tracking and debugging. For example:5960```json JSON61{62  "type": "error",63  "error": {64    "type": "not_found_error",65    "message": "The requested resource could not be found."66  },67  "request_id": "req_011CSHoEeqs5C35K2UUqR7Fy"68}69```7071In accordance with the [versioning](https://platform.claude.com/docs/en/api/versioning) policy, the values within these objects may expand, and it is possible that the `type` values will grow over time.7273## SDK error types7475The official SDKs raise typed exceptions for these errors instead of returning raw JSON, and the class names and namespaces differ by language. For example, a 404 surfaces as `anthropic.NotFoundError` in Python, `Anthropic::Errors::NotFoundError` in Ruby, `com.anthropic.errors.NotFoundException` in Java, and as a single `*anthropic.Error` value (branch on `StatusCode`) in Go. Catch the SDK's typed classes rather than string-matching error messages, handling the most specific classes first. Each SDK page documents its full exception hierarchy:7677* [Python](https://platform.claude.com/docs/en/cli-sdks-libraries/sdks/python#handling-errors) · [TypeScript](https://platform.claude.com/docs/en/cli-sdks-libraries/sdks/typescript#handling-errors) · [C#](https://platform.claude.com/docs/en/cli-sdks-libraries/sdks/csharp#error-handling) · [Go](https://platform.claude.com/docs/en/cli-sdks-libraries/sdks/go#error-handling) · [Java](https://platform.claude.com/docs/en/cli-sdks-libraries/sdks/java#error-handling) · [PHP](https://platform.claude.com/docs/en/cli-sdks-libraries/sdks/php#error-handling) · [Ruby](https://platform.claude.com/docs/en/cli-sdks-libraries/sdks/ruby#handling-errors)7879## Request ID8081Every API response includes a unique `request-id` header. This header contains a value such as `req_018EeWyXxfu5pfWkrYcMdjWG`. The same identifier appears as the `request_id` field in [error response bodies](https://platform.claude.com/docs/en/api/errors#error-shapes). When contacting support about a specific request, include this ID to help quickly resolve your issue.8283On [Claude Platform on AWS](https://platform.claude.com/docs/en/build-with-claude/claude-platform-on-aws), responses include two request IDs: the AWS request ID (`x-amzn-requestid`, primary, indexed in CloudTrail) and the Anthropic request ID (`request-id`, secondary). Use the AWS request ID for CloudTrail lookups and the Anthropic request ID for Anthropic support tickets.8485The Python and TypeScript SDKs expose the request ID as a `_request_id` property on top-level response objects. The C#, Go, Java, and PHP SDKs expose it through their raw-response accessors, and the Ruby SDK through [middleware](https://platform.claude.com/docs/en/cli-sdks-libraries/middleware). The same mechanisms, along with `with_raw_response` in Python and `.withResponse()` in TypeScript, read any other [response header](https://platform.claude.com/docs/en/api/overview#response-headers) too, such as `anthropic-organization-id` and [`anthropic-workspace-id`](https://platform.claude.com/docs/en/manage-claude/workspaces#identify-the-workspace-behind-an-api-response). On Claude Platform on AWS, use the raw-response accessor to read the AWS request ID (`x-amzn-requestid`) as well:8687<CodeGroup>88  ```bash cURL89  # Print the response headers (including request-id); discard the body90  curl -sS -D - -o /dev/null https://api.anthropic.com/v1/messages \91    -H "x-api-key: $ANTHROPIC_API_KEY" \92    -H "anthropic-version: 2023-06-01" \93    -H "content-type: application/json" \94    -d '{95      "model": "claude-sonnet-5",96      "max_tokens": 1024,97      "messages": [{"role": "user", "content": "Hello, Claude"}]98    }'99  ```100101  ```bash CLI102  # The request-id header is printed to stderr with --debug:103  ant --debug messages create \104    --model claude-sonnet-5 \105    --max-tokens 1024 \106    --message '{role: user, content: "Hello, Claude"}'107  ```108109  ```python Python110  client = anthropic.Anthropic()111112  message = client.messages.create(113      model="claude-sonnet-5",114      max_tokens=1024,115      messages=[{"role": "user", "content": "Hello, Claude"}],116  )117  print(f"Request ID: {message._request_id}")118  ```119120  ```typescript TypeScript121  const client = new Anthropic();122123  const message = await client.messages.create({124    model: "claude-sonnet-5",125    max_tokens: 1024,126    messages: [{ role: "user", content: "Hello, Claude" }]127  });128  console.log("Request ID:", message._request_id);129  ```130131  ```csharp C#132  AnthropicClient client = new();133134  using var response = await client.WithRawResponse.Messages.Create(new MessageCreateParams135  {136      Model = Model.ClaudeSonnet5,137      MaxTokens = 1024,138      Messages = [new() { Role = Role.User, Content = "Hello, Claude" }]139  });140  Console.WriteLine($"Request ID: {response.RequestID}");141  ```142143  ```go Go144  client := anthropic.NewClient()145146  var response *http.Response147  _, err := client.Messages.New(148  	context.Background(),149  	anthropic.MessageNewParams{150  		Model:     anthropic.ModelClaudeSonnet5,151  		MaxTokens: 1024,152  		Messages: []anthropic.MessageParam{153  			anthropic.NewUserMessage(anthropic.NewTextBlock("Hello, Claude")),154  		},155  	},156  	option.WithResponseInto(&response),157  )158  if err != nil {159  	panic(err)160  }161162  fmt.Println("Request ID:", response.Header.Get("request-id"))163  ```164165  ```java Java166  import com.anthropic.client.AnthropicClient;167  import com.anthropic.client.okhttp.AnthropicOkHttpClient;168  import com.anthropic.core.http.HttpResponseFor;169  import com.anthropic.models.messages.Message;170  import com.anthropic.models.messages.MessageCreateParams;171  import com.anthropic.models.messages.Model;172173  void main() {174      AnthropicClient client = AnthropicOkHttpClient.fromEnv();175176      HttpResponseFor<Message> response = client.messages().withRawResponse().create(177          MessageCreateParams.builder()178              .model(Model.CLAUDE_SONNET_5)179              .maxTokens(1024)180              .addUserMessage("Hello, Claude")181              .build()182      );183184      IO.println("Request ID: " + response.requestId().orElse(null));185  }186  ```187188  ```php PHP189  $client = new Client();190191  $response = $client->messages->raw->create([192      'model' => 'claude-sonnet-5',193      'maxTokens' => 1024,194      'messages' => [['role' => 'user', 'content' => 'Hello, Claude']],195  ]);196  echo 'Request ID: ' . $response->getHeaderLine('request-id') . "\n";197  ```198199  ```ruby Ruby200  client = Anthropic::Client.new201202  # Read response headers in per-request middleware, which receives the203  # raw HTTP response before the SDK parses it204  request_id = nil205  read_request_id = lambda do |request, call_next|206    response = call_next.call(request)207    # Keys in response.headers are lowercase208    request_id = response.headers["request-id"]209    response210  end211212  client.messages.create(213    model: Anthropic::Model::CLAUDE_SONNET_5,214    max_tokens: 1024,215    messages: [{ role: "user", content: "Hello, Claude" }],216    request_options: { middleware: [read_request_id] }217  )218  puts "Request ID: #{request_id}"219  ```220221  ```python Python (Claude Platform on AWS)222  from anthropic import AnthropicAWS223224  client = AnthropicAWS(aws_region="us-west-2")225226  response = client.messages.with_raw_response.create(227      model="claude-opus-4-8",228      max_tokens=1024,229      messages=[{"role": "user", "content": "Hello, Claude"}],230  )231  print(f"AWS request ID: {response.headers.get('x-amzn-requestid')}")232  message = response.parse()233  print(f"Anthropic request ID: {message._request_id}")234  ```235236  ```typescript TypeScript (Claude Platform on AWS)237  import AnthropicAws from "@anthropic-ai/aws-sdk";238239  const client = new AnthropicAws({ awsRegion: "us-west-2" });240241  const { response: raw, request_id } = await client.messages242    .create({243      model: "claude-opus-4-8",244      max_tokens: 1024,245      messages: [{ role: "user", content: "Hello, Claude" }]246    })247    .withResponse();248  console.log("AWS request ID:", raw.headers.get("x-amzn-requestid"));249  console.log("Anthropic request ID:", request_id);250  ```251</CodeGroup>252253For Claude Platform on AWS request-ID examples in other languages, see [Request IDs](https://platform.claude.com/docs/en/build-with-claude/claude-platform-on-aws#request-ids).254255## Long requests256257<Warning>258  Consider using the [streaming Messages API](https://platform.claude.com/docs/en/build-with-claude/streaming) or [Message Batches API](https://platform.claude.com/docs/en/api/messages/batches/create) for long-running requests, especially those over 10 minutes.259</Warning>260261Avoid setting a large `max_tokens` value without using the [streaming Messages API](https://platform.claude.com/docs/en/build-with-claude/streaming) or [Message Batches API](https://platform.claude.com/docs/en/api/messages/batches/create):262263* Some networks may drop idle connections after a variable period of time, which can cause the request to fail or time out without receiving a response from Anthropic.264* Networks differ in reliability. The [Message Batches API](https://platform.claude.com/docs/en/api/messages/batches/create) can help you manage the risk of network issues by allowing you to poll for results rather than requiring an uninterrupted network connection.265266If you are building a direct API integration, setting a [TCP socket keep-alive](https://tldp.org/HOWTO/TCP-Keepalive-HOWTO/programming.html) can reduce the impact of idle connection timeouts on some networks.267268The [SDKs](https://platform.claude.com/docs/en/cli-sdks-libraries/overview) validate that your non-streaming Messages API requests are not expected to exceed a 10-minute timeout. They also set a socket option for TCP keep-alive.269270If you don't need to process events incrementally, the SDKs can consume the stream for you and return the complete `Message` object, identical to what a non-streaming call returns:271272<CodeGroup>273  ```bash cURL274  # Raw SSE output requires handling events; there is no single-command way275  # to accumulate the final message with curl. Use the SDK examples instead.276  ```277278  ```bash CLI279  # The CLI streams events; --format jsonl emits one event per line280  ant messages create --stream --format jsonl <<'YAML'281  model: claude-sonnet-5282  max_tokens: 128000283  messages:284    - role: user285      content: Write a detailed analysis...286  YAML287  ```288289  ```python Python290  client = anthropic.Anthropic()291292  with client.messages.stream(293      max_tokens=128000,294      messages=[{"role": "user", "content": "Write a detailed analysis..."}],295      model="claude-sonnet-5",296  ) as stream:297      message = stream.get_final_message()298299  print(next(block.text for block in message.content if block.type == "text"))300  ```301302  ```typescript TypeScript303  const client = new Anthropic();304305  const stream = client.messages.stream({306    max_tokens: 128000,307    messages: [{ role: "user", content: "Write a detailed analysis..." }],308    model: "claude-sonnet-5"309  });310311  const message = await stream.finalMessage();312  const textBlock = message.content.find((block) => block.type === "text");313  if (textBlock && textBlock.type === "text") {314    console.log(textBlock.text);315  }316  ```317318  ```csharp C#319  AnthropicClient client = new();320321  var parameters = new MessageCreateParams322  {323      Model = Model.ClaudeSonnet5,324      MaxTokens = 128000,325      Messages = [new() { Role = Role.User, Content = "Write a detailed analysis..." }]326  };327328  var message = await client.Messages.CreateStreaming(parameters).Aggregate();329  Console.WriteLine(message);330  ```331332  ```go Go333  client := anthropic.NewClient()334335  stream := client.Messages.NewStreaming(context.TODO(), anthropic.MessageNewParams{336  	Model:     anthropic.ModelClaudeSonnet5,337  	MaxTokens: 128000,338  	Messages: []anthropic.MessageParam{339  		anthropic.NewUserMessage(anthropic.NewTextBlock("Write a detailed analysis...")),340  	},341  })342343  message := anthropic.Message{}344  for stream.Next() {345  	event := stream.Current()346  	if err := message.Accumulate(event); err != nil {347  		log.Fatal(err)348  	}349  }350  if err := stream.Err(); err != nil {351  	log.Fatal(err)352  }353354  for _, block := range message.Content {355  	if textBlock, ok := block.AsAny().(anthropic.TextBlock); ok {356  		fmt.Println(textBlock.Text)357  		break358  	}359  }360  ```361362  ```java Java363  import com.anthropic.client.AnthropicClient;364  import com.anthropic.client.okhttp.AnthropicOkHttpClient;365  import com.anthropic.helpers.MessageAccumulator;366  import com.anthropic.models.messages.ContentBlock;367  import com.anthropic.models.messages.Message;368  import com.anthropic.models.messages.MessageCreateParams;369  import com.anthropic.models.messages.Model;370371  void main() {372      AnthropicClient client = AnthropicOkHttpClient.fromEnv();373374      MessageCreateParams params = MessageCreateParams.builder()375          .model(Model.CLAUDE_SONNET_5)376          .maxTokens(128000L)377          .addUserMessage("Write a detailed analysis...")378          .build();379380      MessageAccumulator accumulator = MessageAccumulator.create();381      try (var streamResponse = client.messages().createStreaming(params)) {382          streamResponse.stream().forEach(accumulator::accumulate);383      }384385      Message message = accumulator.message();386      message.content().stream()387              .filter(ContentBlock::isText)388              .findFirst()389              .flatMap(ContentBlock::text)390              .ifPresent(textBlock -> IO.println(textBlock.text()));391  }392  ```393394  ```php PHP395  use Anthropic\Lib\Streaming\MessageAccumulator;396397  $client = new Client();398399  $stream = $client->messages->createStream(400      model: 'claude-sonnet-5',401      maxTokens: 128000,402      messages: [['role' => 'user', 'content' => 'Write a detailed analysis...']],403  );404405  $accumulator = MessageAccumulator::forMessages();406  foreach ($stream as $event) {407      $accumulator->accumulate($event);408  }409410  echo array_find($accumulator->message()->content, static fn ($block): bool => $block->type === 'text')->text;411  ```412413  ```ruby Ruby414  client = Anthropic::Client.new415416  message = client.messages.stream(417    model: "claude-sonnet-5",418    max_tokens: 128000,419    messages: [{ role: "user", content: "Write a detailed analysis..." }]420  ).accumulated_message421422  puts message.content.find { it.type == :text }.text423  ```424</CodeGroup>425426See [Streaming Messages](https://platform.claude.com/docs/en/build-with-claude/streaming#get-the-final-message-without-handling-events) for more details.427428## Common validation errors429430### Prefill not supported431432Claude 4.6 and later models and [Claude Mythos Preview](https://anthropic.com/glasswing) do not support prefilling assistant messages. Sending a request with a prefilled last assistant message to any of these models returns a 400 `invalid_request_error`:433434```json435{436  "type": "error",437  "error": {438    "type": "invalid_request_error",439    "message": "This model does not support assistant message prefill. The conversation must end with a user message."440  }441}442```443444Use [structured outputs](https://platform.claude.com/docs/en/build-with-claude/structured-outputs) on models that support it, system prompt instructions, or [`output_config.format`](https://platform.claude.com/docs/en/build-with-claude/structured-outputs#json-outputs) instead.445446### Thinking blocks cannot be modified447448If the most recent assistant message contains `thinking` or `redacted_thinking` blocks that were edited, reordered, filtered out, or reconstructed before being sent back to the API, the request returns a 400 `invalid_request_error`. The error message starts with the position of the offending block (for example, `messages.1.content.0`) and contains:449450```text wrap451`thinking` or `redacted_thinking` blocks in the latest assistant message cannot be modified. These blocks must remain as they were in the original response.452```453454With tool use, every `thinking` and `redacted_thinking` block from the assistant turn must be passed back exactly as received, including blocks whose `thinking` field is empty. Pass thinking blocks back unchanged, and if your application filters content blocks by type before resending, include both `thinking` and `redacted_thinking`. See [Troubleshooting thinking](https://platform.claude.com/docs/en/build-with-claude/thinking-troubleshooting#error-thinking-blocks-modified), [Preserving thinking blocks](https://platform.claude.com/docs/en/build-with-claude/thinking#preserving-thinking-blocks), and [Preserved thinking](https://platform.claude.com/docs/en/build-with-claude/thinking#preserved-thinking).455456### Extended thinking not supported457458Claude 4.7 and later models have removed extended thinking. Sending `thinking: {"type": "enabled"}` to any of these models returns a 400 `invalid_request_error`:459460```text wrap461"thinking.type.enabled" is not supported for this model. Use "thinking.type.adaptive" and "output_config.effort" to control thinking behavior.462```463464Use [adaptive thinking](https://platform.claude.com/docs/en/build-with-claude/thinking) instead. [Migrating to adaptive thinking](https://platform.claude.com/docs/en/build-with-claude/extended-thinking#migrating-to-adaptive-thinking) shows the parameter mapping, and [Troubleshooting thinking](https://platform.claude.com/docs/en/build-with-claude/thinking-troubleshooting#error-thinking-type-enabled) covers the symptom-first fix.465466### Adaptive thinking not supported467468Models that support only extended thinking (Claude 4.5 and earlier models) reject `thinking: {"type": "adaptive"}` with a 400 `invalid_request_error`:469470```text wrap471adaptive thinking is not supported on this model472```473474Use `thinking: {"type": "enabled", "budget_tokens": N}` on these models; see [Extended thinking](https://platform.claude.com/docs/en/build-with-claude/extended-thinking) for the configuration and [Troubleshooting thinking](https://platform.claude.com/docs/en/build-with-claude/thinking-troubleshooting#error-thinking-type-adaptive) for the symptom-first fix.475476### Thinking cannot be disabled477478On Claude Fable 5.1, [Claude Mythos 5.1](https://anthropic.com/glasswing), Claude Fable 5, [Claude Mythos 5](https://anthropic.com/glasswing), and [Claude Mythos Preview](https://anthropic.com/glasswing), thinking is always on. Sending `thinking: {"type": "disabled"}` to any of these models returns a 400 `invalid_request_error`. On all of these models except Claude Mythos Preview, the message reads:479480```text wrap481"thinking.type.disabled" is not supported for this model. Use "thinking.type.adaptive" and "output_config.effort" to control thinking behavior.482```483484On Claude Mythos Preview, the only one of these models that accepts extended thinking, the message reads:485486```text wrap487"thinking.type.disabled" is not supported for this model. Thinking defaults to adaptive mode when not specified; use "thinking.type.enabled" with "budget_tokens" for extended thinking.488```489490Omit the `thinking` parameter and the request runs with adaptive thinking. To keep thinking content out of responses without turning thinking off, set `display: "omitted"` on the thinking configuration. See [Troubleshooting thinking](https://platform.claude.com/docs/en/build-with-claude/thinking-troubleshooting#error-thinking-type-disabled).491492### Forced tool use not supported493494Claude Fable 5.1 and [Claude Mythos 5.1](https://anthropic.com/glasswing) don't support forced tool use. Sending `tool_choice: {"type": "any"}` or `tool_choice: {"type": "tool", "name": "..."}` to either model, including on the [token counting endpoint](https://platform.claude.com/docs/en/build-with-claude/token-counting), returns a 400 `invalid_request_error`:495496```text wrap497tool_choice: type "tool" and "any" are not supported for this model.498```499500`tool_choice: {"type": "auto"}` (the default) and `{"type": "none"}` are accepted. Use `auto` with [strict tool use](https://platform.claude.com/docs/en/agents-and-tools/tool-use/strict-tool-use) to keep tool inputs schema-valid, or [structured outputs](https://platform.claude.com/docs/en/build-with-claude/structured-outputs) when you need the response itself in a fixed JSON shape. See [Forcing tool use](https://platform.claude.com/docs/en/agents-and-tools/tool-use/define-tools#forcing-tool-use).501502### Thinking block no longer matches the conversation503504On Claude Fable 5.1, the API accepts a replayed thinking block only while the `system` prompt, `tools`, and messages that preceded it are unchanged. For new accounts created on or after August 31, 2026, and for any request that sets `thinking.block_binding.prefix_mismatch_behavior` to `"error"`, a replayed block whose earlier history changed is rejected with a 400 `invalid_request_error` (with `"drop_block"`, the API drops the block and the request succeeds). The message starts with the position of the first failing block:505506```text wrap507messages.{i}.content.{j}: Invalid `signature` in `thinking` block. The block is bound to a different conversation. Remove the block, or set `thinking.block_binding.prefix_mismatch_behavior` to "drop_block".508```509510Without the `thinking-binding-controls-2026-08-01` beta header the message also names that header. Keep the conversation history append-only, or send the beta header with `prefix_mismatch_behavior: "drop_block"` to drop the block and continue. A block from a model the target model can't read is dropped rather than rejected. See [Keeping the prefix unchanged](https://platform.claude.com/docs/en/build-with-claude/preserved-thinking#prefix-check) and [Troubleshooting thinking](https://platform.claude.com/docs/en/build-with-claude/thinking-troubleshooting#error-thinking-block-signature).511512Sending `thinking.block_binding` without the `thinking-binding-controls-2026-08-01` [beta header](https://platform.claude.com/docs/en/api/beta-headers) returns a 400 `invalid_request_error` whose message ends in:513514```text wrap515block_binding: Extra inputs are not permitted516```517518Add the header, or remove the field.519520### Outbound web identity federation disabled (Claude Platform on AWS)521522If every request to [Claude Platform on AWS](https://platform.claude.com/docs/en/build-with-claude/claude-platform-on-aws) returns `"Outbound web identity federation is disabled for your account"`, run `aws iam enable-outbound-web-identity-federation` once per AWS account. See [Enable outbound web identity federation](https://platform.claude.com/docs/en/build-with-claude/claude-platform-on-aws#enable-outbound-web-identity-federation) for details.523524## Next steps525526<CardGroup cols={3}>527  <Card title="Troubleshooting thinking" icon="wrench" href="https://platform.claude.com/docs/en/build-with-claude/thinking-troubleshooting">528    Symptom-first fixes for thinking configuration 400 errors, empty thinking blocks, and `max_tokens` stops.529  </Card>530531  <Card title="Rate limits" icon="gauge" href="https://platform.claude.com/docs/en/api/rate-limits">532    To mitigate misuse and manage capacity on the API, limits are in place on how much an organization can use the Claude API.533  </Card>534535  <Card title="Streaming messages" icon="lightning" href="https://platform.claude.com/docs/en/build-with-claude/streaming">536    Stream Messages API responses incrementally with server-sent events, including text, tool use, and extended thinking deltas.537  </Card>538</CardGroup>539
Files main
Menu

Theme