---
type: "firecrawl-provider"
description: "Official Claude Developer Platform documentation (platform.claude.com/docs, formerly docs.anthropic.com): search the docs, read any page or one section as clean Markdown, the current model lineup, one model's specs, token pricing, model deprecations and retirements, dated API release notes, and Claude API endpoint references. Public Markdown twins and Anthropic's public docs MCP server; no API key."
use_when: "Official Claude Developer Platform documentation (platform.claude.com/docs, formerly docs.anthropic.com): search the docs, read any page or one section as clean Markdown, the current model lineup, one model's specs, token pricing, model deprecations and retirements, dated API release notes, and Claude API endpoint references. Public Markdown twins and Anthropic's public docs MCP server; no API key."
categories: "AI models"
capabilities: 8
credits_per_call: 5
---
# Claude Developer Platform docs on Firecrawl Alexandria

Official Claude Developer Platform documentation (platform.claude.com/docs, formerly docs.anthropic.com): search the docs, read any page or one section as clean Markdown, the current model lineup, one model's specs, token pricing, model deprecations and retirements, dated API release notes, and Claude API endpoint references. Public Markdown twins and Anthropic's public docs MCP server; no API key.

- Categories: AI models
- Category index: [AI models category](https://firecrawl.dev/alexandria/agents/categories/ai-models)
- Provider key: `platform-claude-com`
- Access: Firecrawl credits
- Cost: 5 credits per call

## More

- [Human guide](https://firecrawl.dev/app/alexandria/platform-claude-com)
- [OpenAPI spec](https://firecrawl.dev/alexandria/agents/providers/platform-claude-com/openapi.json)

## Capabilities

- [Get api endpoint](https://firecrawl.dev/alexandria/agents/providers/platform-claude-com/developer-docs/get_api_endpoint): One Claude API endpoint's reference from Anthropic's docs MCP server (`get_api_endpoint`): title, summary and description, every top-level path, query, header and body parameter (name, type, required, deprecated, description), the response object's type and top-level fields, the curl example request and example response, and the full reference Markdown (truncate with `max_chars`). Unknown endpoints are `not_found`.
- [Get model](https://firecrawl.dev/alexandria/agents/providers/platform-claude-com/developer-docs/get_model): One Claude model's official page (`/docs/en/models/{slug}/overview.md`) by display name, API ID (dated or not), Bedrock/Vertex ID or slug: API ID, release date and lifecycle tag, status, retirement, platforms and platform IDs, input/output price per million tokens, context window, max output, thinking, default effort, modalities, cutoffs, the full pricing, capabilities and availability tables, the page's Good to know notes (such as sampling parameters that return 400), outline and Markdown. Unknown models are `not_found`.
- [Get page](https://firecrawl.dev/alexandria/agents/providers/platform-claude-com/developer-docs/get_page): One Claude Developer Platform docs page as clean Markdown from its official `.md` twin (`/docs/en/{path}.md`): title, description, URL, an outline of every heading (level, text, anchor), and the content. Pass `section` (heading text or anchor, case-insensitive) to get only that heading's subtree; an unknown section is `not_found`. Accepts a docs path or a full docs URL from platform.claude.com, docs.claude.com or docs.anthropic.com, with or without `.md` or `#anchor` (an anchor selects that section).
- [List models](https://firecrawl.dev/alexandria/agents/providers/platform-claude-com/developer-docs/list_models): The current Claude model lineup from the official models overview (`/docs/en/models/overview.md`): one record per current model with API ID, alias, description, pricing (USD per million input and output tokens, plus the raw text), context window and max output in tokens, thinking mode, default effort, knowledge and training cutoffs, retirement commitment, platform model IDs and model page; plus the legacy models still available and the table's footnotes.
- [Model deprecations](https://firecrawl.dev/alexandria/agents/providers/platform-claude-com/developer-docs/model_deprecations): Model lifecycle from the official deprecations page (`/docs/en/about-claude/model-deprecations.md`): per API model ID its state (Active, Deprecated, Retired), deprecation date, retirement date or not-sooner-than commitment, recommended replacement and announcement date; every deprecation announcement row; and deprecated API parameters with their behavior (for example sampling parameters that return 400 on newer models). Filter by model substring.
- [Model pricing](https://firecrawl.dev/alexandria/agents/providers/platform-claude-com/developer-docs/model_pricing): Official Claude API token prices from the pricing page (`/docs/en/about-claude/pricing.md`): per model, base input, 5-minute and 1-hour cache writes, cache hits, output, and Batch API input/output, each in USD per million tokens as a number plus the raw text, with tier notes (such as prompt-length tiers) and footnote markers; plus fast mode prices and the table footnotes. Filter by model substring.
- [Release notes](https://firecrawl.dev/alexandria/agents/providers/platform-claude-com/developer-docs/release_notes): Dated Claude Platform release notes (Claude API, SDKs, Console) from `/docs/en/release-notes/overview.md`, newest first: each entry's ISO date and its items as Markdown. Filter with `since` (YYYY-MM-DD, inclusive) and `q` (keeps only items containing the text); page with `limit` and `offset` (`next_offset` continues).
- [Search docs](https://firecrawl.dev/alexandria/agents/providers/platform-claude-com/developer-docs/search_docs): Search the official Claude Developer Platform documentation (guides and the API reference) through the docs' own search, the `search_docs` tool of Anthropic's public docs MCP server. Returns ranked results with title, breadcrumb, kind, URL, docs path, description, endpoint (for API methods) and a snippet, in the vendor's ranking order.

## 1. Choose this provider when

Official Claude Developer Platform documentation (platform.claude.com/docs, formerly docs.anthropic.com): search the docs, read any page or one section as clean Markdown, the current model lineup, one model's specs, token pricing, model deprecations and retirements, dated API release notes, and Claude API endpoint references. Public Markdown twins and Anthropic's public docs MCP server; no API key.

## 2. Minimal request

Call `POST https://api.firecrawl.dev/v2/scrape` with `{ alexandria: { provider, capability, options } }`. For a batch, send `{ alexandria: [...] }` with up to 10 calls.

```json
{
  "provider": "platform-claude-com",
  "capability": "developer-docs/get_api_endpoint",
  "options": {
    "max_chars": 400,
    "method": "GET",
    "path": "/v1/models/{model_id}"
  }
}
```

## 3. Add provider options

Use only the options needed for the task:

- `max_chars` (number): Truncate `content` to this many characters; the whole text is returned when omitted. Example: `200`
- `method` (string, required): HTTP method. Example: `GET`
- `path` (string, required): API path with placeholders as documented, e.g. `/v1/messages` or `/v1/models/{model_id}`. Example: `/v1/messages`

## 4. Request through your preferred interface

### JavaScript

```javascript
const result = await firecrawl.scrape({
  alexandria: {
    provider: "platform-claude-com",
    capability: "developer-docs/get_api_endpoint",
    options: {
      max_chars: 400,
      method: "GET",
      path: "/v1/models/{model_id}",
    },
  },
});
```

### Python

```python
result = firecrawl.scrape_alexandria({
  "provider": "platform-claude-com",
  "capability": "developer-docs/get_api_endpoint",
  "options": {
    "max_chars": 400,
    "method": "GET",
    "path": "/v1/models/{model_id}"
  }
})
```

### cURL

```sh
curl https://api.firecrawl.dev/v2/scrape \
  -H "Authorization: Bearer $FIRECRAWL_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
  "alexandria": {
    "provider": "platform-claude-com",
    "capability": "developer-docs/get_api_endpoint",
    "options": {
      "max_chars": 400,
      "method": "GET",
      "path": "/v1/models/{model_id}"
    }
  }
}'
```

### CLI

```sh
firecrawl scrape 'platform-claude-com/developer-docs/get_api_endpoint' \
  --options '{"max_chars":400,"method":"GET","path":"/v1/models/{model_id}"}'
```


### MCP

Call the FCX MCP retrieve tool with this object:

```json
{
  "provider": "platform-claude-com",
  "capability": "developer-docs/get_api_endpoint",
  "options": {
    "max_chars": 400,
    "method": "GET",
    "path": "/v1/models/{model_id}"
  }
}
```

Ask for only the returned fields needed by the task.

## 5. Full request shape

```json
{
  "provider": "platform-claude-com",
  "capability": "developer-docs/get_api_endpoint",
  "options": {
    "max_chars": 400,
    "method": "GET",
    "path": "/v1/models/{model_id}"
  }
}
```

## 6. Response data

The response includes `success`, `provider`, `capability`, `creditsCost` and `data`. This example shows the provider payload in `data`:

```json
{
  "chars": 11656,
  "content": "# Get a Model\n\n**GET** `/v1/models/{model_id}`\n\nGet a specific model.\n\nThe Models API response can be used to determine information about a specific model or resolve a model alias to a model ID.\n\n## Path parameters\n\n- `model_id: string`\n\n  Model identifier or alias.\n\n## Headers\n\n- `\"anthropic-version\": optional string`\n\n  The version of the Claude API you want to use.\n\n  Read more about versioning",
  "description": "Get a specific model.\n\nThe Models API response can be used to determine information about a specific model or resolve a model alias to a model ID.",
  "example_request": "curl https://api.anthropic.com/v1/models/$MODEL_ID \\\n    -H 'anthropic-version: 2023-06-01' \\\n    -H \"X-Api-Key: $ANTHROPIC_API_KEY\"",
  "example_response": "{\n  \"id\": \"claude-opus-5-5\",\n  \"capabilities\": {\n    \"batch\": {\n      \"supported\": true\n    },\n    \"citations\": {\n      \"supported\": true\n    },\n    \"code_execution\": {\n      \"supported\": true\n    },\n    \"context_management\": {\n      \"clear_thinking_20251015\": {\n        \"supported\": true\n      },\n      \"clear_tool_uses_20250919\": {\n        \"supported\": true\n      },\n      \"compact_20260112\": {\n        \"supported\": true\n      },\n      \"supported\": true\n    },\n    \"effort\": {\n      \"high\": {\n        \"supported\": true\n      },\n      \"low\": {\n        \"supported\": true\n      },\n      \"max\": {\n        \"supported\": true\n      },\n      \"medium\": {\n        \"supported\": true\n      },\n      \"supported\": true,\n      \"xhigh\": {\n        \"supported\": true\n      }\n    },\n    \"image_input\": {\n      \"supported\": true\n    },\n    \"pdf_input\": {\n      \"supported\": true\n    },\n    \"server_tools\": {\n      \"code_execution\": {\n        \"supported\": true\n      },\n      \"supported\": true,\n      \"web_search\": {\n        \"supported\": true\n      }\n    },\n    \"structured_outputs\": {\n      \"supported\": true\n    },\n    \"thinking\": {\n      \"supported\": true,\n      \"types\": {\n        \"adaptive\": {\n          \"supported\": true\n        },\n        \"between_tools\": {\n          \"supported\": true\n        },\n        \"disabled\": {\n          \"supported\": true\n        },\n        \"enabled\": {\n          \"supported\": true\n        }\n      }\n    }\n  },\n  \"created_at\": \"2026-09-21T00:00:00Z\",\n  \"deprecated_at\": \"2019-12-27T18:11:19.117Z\",\n  \"display_name\": \"Claude Opus 5.5\",\n  \"lifecycle\": \"active\",\n  \"line\": \"haiku\",\n  \"max_input_tokens\": 0,\n  \"max_tokens\": 0,\n  \"retires_at\": \"2019-12-27T18:11:19.117Z\",\n  \"type\": \"model\"\n}",
  "example_response_status": "200",
  "method": "GET",
  "observed_at_ms": 1792516055058,
  "parameters": [
    {
      "deprecated": false,
      "description": "Model identifier or alias.",
      "location": "path",
      "name": "model_id",
      "required": true,
      "type": "string"
    },
    {
      "deprecated": false,
      "description": "The version of the Claude API you want to use.",
      "location": "header",
      "name": "anthropic-version",
      "required": false,
      "type": "string"
    },
    {
      "deprecated": false,
      "description": "Optional header to select the Workspace for this request. The value is a Workspace ID (for example, wrkspc_011CZkZaBF1tNoB5wlCeusgy).",
      "location": "header",
      "name": "anthropic-workspace-id",
      "required": false,
      "type": "string"
    },
    {
      "deprecated": true,
      "description": "Optional header to specify the beta version(s) you want to use.",
      "location": "header",
      "name": "anthropic-beta",
      "required": false,
      "type": "array of AnthropicBeta"
    }
  ],
  "path": "/v1/models/{model_id}",
  "returns": {
    "fields": [
      {
        "deprecated": false,
        "description": "Object type.",
        "name": "type",
        "required": true,
        "type": "\"model\""
      },
      {
        "deprecated": false,
        "description": "Unique model identifier.",
        "name": "id",
        "required": true,
        "type": "string"
      },
      {
        "deprecated": false,
        "description": "Object mapping capability names to their support details. Keys are always present for all known capabilities.",
        "name": "capabilities",
        "required": true,
        "type": "ModelCapabilities or null"
      },
      {
        "deprecated": false,
        "description": "RFC 3339 datetime string representing the time at which the model was released. May be set to an epoch value if the release date is unknown.",
        "name": "created_at",
        "required": true,
        "type": "string"
      },
      {
        "deprecated": false,
        "description": "RFC 3339 datetime string representing the time of the model's most recent deprecation. Populated for deprecated and retired models; null while the model is active.",
        "name": "deprecated_at",
        "required": true,
        "type": "string or null"
      },
      {
        "deprecated": false,
        "description": "A human-readable name for the model.",
        "name": "display_name",
        "required": true,
        "type": "string"
      },
      {
        "deprecated": false,
        "description": "The model's current lifecycle stage.",
        "name": "lifecycle",
        "required": true,
        "type": "\"active\" or \"deprecated\" or \"retired\""
      },
      {
        "deprecated": false,
        "description": "The model line this model belongs to, such as opus for both Claude Opus 4.5 and Claude Opus 4.6. More lines may be added. null when the model belongs to no line; do not infer a line from the id.",
        "name": "line",
        "required": true,
        "type": "ModelLine or null"
      },
      {
        "deprecated": false,
        "description": "Maximum input context window size in tokens for this model.",
        "name": "max_input_tokens",
        "required": true,
        "type": "number or null"
      },
      {
        "deprecated": false,
        "description": "Maximum value for the max_tokens parameter when using this model.",
        "name": "max_tokens",
        "required": true,
        "type": "number or null"
      },
      {
        "deprecated": false,
        "description": "RFC 3339 datetime string representing the model's currently scheduled retirement date. The schedule can be revised until retirement occurs; null while the model is active or while no retirement is scheduled. A past date on a deprecated model means retirement is overdue, not that it has occurred: lifecycle is the retirement signal.",
        "name": "retires_at",
        "required": true,
        "type": "string or null"
      }
    ],
    "type": "ModelInfo object"
  },
  "source_url": "https://platform.claude.com/docs/mcp",
  "summary": "Get a specific model.",
  "title": "Get a Model",
  "truncated": true,
  "url": "https://platform.claude.com/docs/en/api/models/retrieve"
}
```

## API reference-derived contract

The following capability contract is generated from the same normalized Alexandria API reference exposed in the API spec.

### Get api endpoint

- Capability: `developer-docs/get_api_endpoint`
- Description: One Claude API endpoint's reference from Anthropic's docs MCP server (`get_api_endpoint`): title, summary and description, every top-level path, query, header and body parameter (name, type, required, deprecated, description), the response object's type and top-level fields, the curl example request and example response, and the full reference Markdown (truncate with `max_chars`). Unknown endpoints are `not_found`.
- Instructions: To look up a Claude API endpoint's parameters and response shape, e.g. POST /v1/messages or GET /v1/models/{model_id}.
- Cost: 5 credits per call
- Capability file: [Get api endpoint](https://firecrawl.dev/alexandria/agents/providers/platform-claude-com/developer-docs/get_api_endpoint)

Accepted options:
- `max_chars` (number): Truncate `content` to this many characters; the whole text is returned when omitted. Example: `200`
- `method` (string, required): HTTP method. Example: `GET`
- `path` (string, required): API path with placeholders as documented, e.g. `/v1/messages` or `/v1/models/{model_id}`. Example: `/v1/messages`

Response schema example:
```json
{
  "chars": 11656,
  "content": "# Get a Model\n\n**GET** `/v1/models/{model_id}`\n\nGet a specific model.\n\nThe Models API response can be used to determine information about a specific model or resolve a model alias to a model ID.\n\n## Path parameters\n\n- `model_id: string`\n\n  Model identifier or alias.\n\n## Headers\n\n- `\"anthropic-version\": optional string`\n\n  The version of the Claude API you want to use.\n\n  Read more about versioning",
  "description": "Get a specific model.\n\nThe Models API response can be used to determine information about a specific model or resolve a model alias to a model ID.",
  "example_request": "curl https://api.anthropic.com/v1/models/$MODEL_ID \\\n    -H 'anthropic-version: 2023-06-01' \\\n    -H \"X-Api-Key: $ANTHROPIC_API_KEY\"",
  "example_response": "{\n  \"id\": \"claude-opus-5-5\",\n  \"capabilities\": {\n    \"batch\": {\n      \"supported\": true\n    },\n    \"citations\": {\n      \"supported\": true\n    },\n    \"code_execution\": {\n      \"supported\": true\n    },\n    \"context_management\": {\n      \"clear_thinking_20251015\": {\n        \"supported\": true\n      },\n      \"clear_tool_uses_20250919\": {\n        \"supported\": true\n      },\n      \"compact_20260112\": {\n        \"supported\": true\n      },\n      \"supported\": true\n    },\n    \"effort\": {\n      \"high\": {\n        \"supported\": true\n      },\n      \"low\": {\n        \"supported\": true\n      },\n      \"max\": {\n        \"supported\": true\n      },\n      \"medium\": {\n        \"supported\": true\n      },\n      \"supported\": true,\n      \"xhigh\": {\n        \"supported\": true\n      }\n    },\n    \"image_input\": {\n      \"supported\": true\n    },\n    \"pdf_input\": {\n      \"supported\": true\n    },\n    \"server_tools\": {\n      \"code_execution\": {\n        \"supported\": true\n      },\n      \"supported\": true,\n      \"web_search\": {\n        \"supported\": true\n      }\n    },\n    \"structured_outputs\": {\n      \"supported\": true\n    },\n    \"thinking\": {\n      \"supported\": true,\n      \"types\": {\n        \"adaptive\": {\n          \"supported\": true\n        },\n        \"between_tools\": {\n          \"supported\": true\n        },\n        \"disabled\": {\n          \"supported\": true\n        },\n        \"enabled\": {\n          \"supported\": true\n        }\n      }\n    }\n  },\n  \"created_at\": \"2026-09-21T00:00:00Z\",\n  \"deprecated_at\": \"2019-12-27T18:11:19.117Z\",\n  \"display_name\": \"Claude Opus 5.5\",\n  \"lifecycle\": \"active\",\n  \"line\": \"haiku\",\n  \"max_input_tokens\": 0,\n  \"max_tokens\": 0,\n  \"retires_at\": \"2019-12-27T18:11:19.117Z\",\n  \"type\": \"model\"\n}",
  "example_response_status": "200",
  "method": "GET",
  "observed_at_ms": 1792516055058,
  "parameters": [
    {
      "deprecated": false,
      "description": "Model identifier or alias.",
      "location": "path",
      "name": "model_id",
      "required": true,
      "type": "string"
    },
    {
      "deprecated": false,
      "description": "The version of the Claude API you want to use.",
      "location": "header",
      "name": "anthropic-version",
      "required": false,
      "type": "string"
    },
    {
      "deprecated": false,
      "description": "Optional header to select the Workspace for this request. The value is a Workspace ID (for example, wrkspc_011CZkZaBF1tNoB5wlCeusgy).",
      "location": "header",
      "name": "anthropic-workspace-id",
      "required": false,
      "type": "string"
    },
    {
      "deprecated": true,
      "description": "Optional header to specify the beta version(s) you want to use.",
      "location": "header",
      "name": "anthropic-beta",
      "required": false,
      "type": "array of AnthropicBeta"
    }
  ],
  "path": "/v1/models/{model_id}",
  "returns": {
    "fields": [
      {
        "deprecated": false,
        "description": "Object type.",
        "name": "type",
        "required": true,
        "type": "\"model\""
      },
      {
        "deprecated": false,
        "description": "Unique model identifier.",
        "name": "id",
        "required": true,
        "type": "string"
      },
      {
        "deprecated": false,
        "description": "Object mapping capability names to their support details. Keys are always present for all known capabilities.",
        "name": "capabilities",
        "required": true,
        "type": "ModelCapabilities or null"
      },
      {
        "deprecated": false,
        "description": "RFC 3339 datetime string representing the time at which the model was released. May be set to an epoch value if the release date is unknown.",
        "name": "created_at",
        "required": true,
        "type": "string"
      },
      {
        "deprecated": false,
        "description": "RFC 3339 datetime string representing the time of the model's most recent deprecation. Populated for deprecated and retired models; null while the model is active.",
        "name": "deprecated_at",
        "required": true,
        "type": "string or null"
      },
      {
        "deprecated": false,
        "description": "A human-readable name for the model.",
        "name": "display_name",
        "required": true,
        "type": "string"
      },
      {
        "deprecated": false,
        "description": "The model's current lifecycle stage.",
        "name": "lifecycle",
        "required": true,
        "type": "\"active\" or \"deprecated\" or \"retired\""
      },
      {
        "deprecated": false,
        "description": "The model line this model belongs to, such as opus for both Claude Opus 4.5 and Claude Opus 4.6. More lines may be added. null when the model belongs to no line; do not infer a line from the id.",
        "name": "line",
        "required": true,
        "type": "ModelLine or null"
      },
      {
        "deprecated": false,
        "description": "Maximum input context window size in tokens for this model.",
        "name": "max_input_tokens",
        "required": true,
        "type": "number or null"
      },
      {
        "deprecated": false,
        "description": "Maximum value for the max_tokens parameter when using this model.",
        "name": "max_tokens",
        "required": true,
        "type": "number or null"
      },
      {
        "deprecated": false,
        "description": "RFC 3339 datetime string representing the model's currently scheduled retirement date. The schedule can be revised until retirement occurs; null while the model is active or while no retirement is scheduled. A past date on a deprecated model means retirement is overdue, not that it has occurred: lifecycle is the retirement signal.",
        "name": "retires_at",
        "required": true,
        "type": "string or null"
      }
    ],
    "type": "ModelInfo object"
  },
  "source_url": "https://platform.claude.com/docs/mcp",
  "summary": "Get a specific model.",
  "title": "Get a Model",
  "truncated": true,
  "url": "https://platform.claude.com/docs/en/api/models/retrieve"
}
```

### Get model

- Capability: `developer-docs/get_model`
- Description: One Claude model's official page (`/docs/en/models/{slug}/overview.md`) by display name, API ID (dated or not), Bedrock/Vertex ID or slug: API ID, release date and lifecycle tag, status, retirement, platforms and platform IDs, input/output price per million tokens, context window, max output, thinking, default effort, modalities, cutoffs, the full pricing, capabilities and availability tables, the page's Good to know notes (such as sampling parameters that return 400), outline and Markdown. Unknown models are `not_found`.
- Instructions: For a structured overview of one model (for example Claude Haiku 5.5: pricing, context window, availability, parameter restrictions).
- Cost: 5 credits per call
- Capability file: [Get model](https://firecrawl.dev/alexandria/agents/providers/platform-claude-com/developer-docs/get_model)

Accepted options:
- `max_chars` (number): Truncate `content` to this many characters; the whole text is returned when omitted. Example: `200`
- `model` (string): Model name, API ID or slug, e.g. `Claude Haiku 5.5`, `claude-haiku-5-5`, `claude-haiku-4-5-20251001`, `haiku-5-5`. Example: `Claude Haiku 5.5`
- `url` (string): Alternatively a model page URL, e.g. https://platform.claude.com/docs/en/models/haiku-5-5/overview. Pattern: ^https?://. Example: `<url>`

Response schema example:
```json
{
  "api_id": "claude-haiku-5-5",
  "availability": [
    {
      "feature": "Status",
      "value": "Active (latest)"
    },
    {
      "feature": "Released",
      "value": "October 7, 2026"
    },
    {
      "feature": "Retirement",
      "value": "Not sooner than October 7, 2027"
    },
    {
      "feature": "Platforms",
      "value": "Claude API, Amazon Bedrock, Google Cloud, Microsoft Foundry, Claude Platform on AWS"
    }
  ],
  "capabilities": [
    {
      "feature": "Context window",
      "value": "1M tokens"
    },
    {
      "feature": "Max output",
      "value": "128K tokens"
    },
    {
      "feature": "Max output (Batch API, beta)",
      "value": "300K tokens"
    },
    {
      "feature": "Thinking",
      "value": "Adaptive"
    },
    {
      "feature": "Default effort",
      "value": "medium"
    },
    {
      "feature": "Comparative latency",
      "value": "Fastest"
    },
    {
      "feature": "Input → output",
      "value": "Text and images → text"
    },
    {
      "feature": "Reliable knowledge cutoff",
      "value": "Jun 2026"
    },
    {
      "feature": "Training data cutoff",
      "value": "Jun 2026"
    }
  ],
  "chars": 14372,
  "content": "**Latest.** Released October 7, 2026.\n\nFor high-volume, latency-sensitive tasks such as classification, extraction, and routing\n\nModel ID: `claude-haiku-5-5`\n\nContext window: 1M tokens · Max output: 128K tokens · Input pricing: From $0.10 / MTok · Output pricing: From $0.50 / MTok\n\n[Announcement](https://www.anthropic.com/claude-haiku-5-5) · [What’s new](https://platform.claude.com/docs/en/models/haiku-5-5/whats-new-haiku-5-5) · [Migration guide](https://platform.claude.com/docs/en/models/haiku-5-5/migration-guide)\n\n## Overview\n\nClaude Haiku 5.5 is built for high-volume, latency-sensitive work",
  "context_window_tokens": 1000000,
  "default_effort": "medium",
  "description": "Claude Haiku 5.5 at a glance: what it's for, model IDs on every platform, context window, output limits, pricing, availability, and the guides and resources for building with it.",
  "good_to_know": [
    "Adaptive thinking is on by default. Control thinking depth with the [effort parameter](https://platform.claude.com/docs/en/build-with-claude/effort).",
    "Omit `temperature`, `top_p`, and `top_k`. See [Remove sampling parameters](https://platform.claude.com/docs/en/models/haiku-5-5/migration-guide#remove-sampling-parameters) for the values that return a 400 error.",
    "On the [Message Batches API](https://platform.claude.com/docs/en/build-with-claude/batch-processing#extended-output-beta), Claude Haiku 5.5 supports up to 300k output tokens with the `output-300k-2026-03-24` beta header.",
    "The minimum cacheable prompt length is 512 tokens. See [Prompt caching](https://platform.claude.com/docs/en/build-with-claude/prompt-caching#cache-limitations).",
    "Query limits and capabilities programmatically with the [Models API](https://platform.claude.com/docs/en/api/models/list)."
  ],
  "input_usd_per_mtok": 0.1,
  "knowledge_cutoff": "Jun 2026",
  "markdown_url": "https://platform.claude.com/docs/en/models/haiku-5-5/overview.md",
  "max_output_tokens": 128000,
  "modalities": "Text and images → text",
  "name": "Claude Haiku 5.5",
  "observed_at_ms": 1792516055041,
  "outline": [
    {
      "anchor": "overview",
      "level": 2,
      "text": "Overview"
    },
    {
      "anchor": "how-it-compares",
      "level": 2,
      "text": "How it compares"
    },
    {
      "anchor": "specifications",
      "level": 2,
      "text": "Specifications"
    },
    {
      "anchor": "model-ids",
      "level": 3,
      "text": "Model IDs"
    },
    {
      "anchor": "pricing",
      "level": 3,
      "text": "Pricing"
    },
    {
      "anchor": "capabilities",
      "level": 3,
      "text": "Capabilities"
    },
    {
      "anchor": "availability",
      "level": 3,
      "text": "Availability"
    },
    {
      "anchor": "good-to-know",
      "level": 2,
      "text": "Good to know"
    },
    {
      "anchor": "resources",
      "level": 2,
      "text": "Resources"
    },
    {
      "anchor": "reference",
      "level": 2,
      "text": "Reference"
    }
  ],
  "output_usd_per_mtok": 0.5,
  "platform_ids": [
    {
      "model_id": "claude-haiku-5-5",
      "platform": "Claude API"
    },
    {
      "model_id": "anthropic.claude-haiku-5-5",
      "platform": "Amazon Bedrock"
    },
    {
      "model_id": "claude-haiku-5-5",
      "platform": "Google Cloud"
    },
    {
      "model_id": "claude-haiku-5-5",
      "platform": "Microsoft Foundry"
    },
    {
      "model_id": "claude-haiku-5-5",
      "platform": "Claude Platform on AWS"
    }
  ],
  "platforms": [
    "Claude API",
    "Amazon Bedrock",
    "Google Cloud",
    "Microsoft Foundry",
    "Claude Platform on AWS"
  ],
  "pricing": [
    {
      "feature": "Input",
      "value": "$0.10 / MTok for prompts up to 100,000 tokens; $0.50 / MTok for prompts over 100,000 tokens"
    },
    {
      "feature": "Output",
      "value": "$0.50 / MTok for prompts up to 100,000 tokens; $2.50 / MTok for prompts over 100,000 tokens"
    },
    {
      "feature": "5m cache write",
      "value": "$0.125 / MTok for prompts up to 100,000 tokens; $0.625 / MTok for prompts over 100,000 tokens"
    },
    {
      "feature": "1h cache write",
      "value": "$0.20 / MTok for prompts up to 100,000 tokens; $1 / MTok for prompts over 100,000 tokens"
    },
    {
      "feature": "Cache read",
      "value": "$0.01 / MTok for prompts up to 100,000 tokens; $0.05 / MTok for prompts over 100,000 tokens"
    },
    {
      "feature": "Batch API",
      "value": "50% discount on input and output"
    }
  ],
  "released": "October 7, 2026",
  "released_date": "2026-10-07",
  "retirement": "Not sooner than October 7, 2027",
  "slug": "haiku-5-5",
  "status": "Active (latest)",
  "summary": "For high-volume, latency-sensitive tasks such as classification, extraction, and routing",
  "tag": "Latest",
  "thinking": "Adaptive",
  "training_data_cutoff": "Jun 2026",
  "truncated": true,
  "url": "https://platform.claude.com/docs/en/models/haiku-5-5/overview"
}
```

### Get page

- Capability: `developer-docs/get_page`
- Description: One Claude Developer Platform docs page as clean Markdown from its official `.md` twin (`/docs/en/{path}.md`): title, description, URL, an outline of every heading (level, text, anchor), and the content. Pass `section` (heading text or anchor, case-insensitive) to get only that heading's subtree; an unknown section is `not_found`. Accepts a docs path or a full docs URL from platform.claude.com, docs.claude.com or docs.anthropic.com, with or without `.md` or `#anchor` (an anchor selects that section).
- Instructions: To read official guidance such as vision or PDF input limits, thinking and effort, or a model's migration guide, ideally one section at a time.
- Cost: 5 credits per call
- Capability file: [Get page](https://firecrawl.dev/alexandria/agents/providers/platform-claude-com/developer-docs/get_page)

Accepted options:
- `max_chars` (number): Truncate `content` to this many characters; the whole text is returned when omitted. Example: `200`
- `path` (string): Docs path such as `build-with-claude/vision` (a `/docs/en/` prefix, `.md` suffix or `#anchor` is accepted). Example: `build-with-claude/vision`
- `section` (string): Heading text or anchor to return only that section, e.g. `Image limits and costs`. Example: `Image limits and costs`
- `url` (string): Alternatively a docs URL, e.g. https://platform.claude.com/docs/en/build-with-claude/pdf-support#check-pdf-requirements. Pattern: ^https?://. Example: `<url>`

Response schema example:
```json
{
  "chars": 8301,
  "content": "## Image limits and costs\n\n### Request limits\n\nThe maximum number of images per message or request is:\n\n* 20 per message on [claude.ai](https://claude.ai/).\n* 100 per request on the API, for models with a 200k-token context window.\n* 600 per request on the API, for all other models.\n\nThe maximum dimensions per image are 8000x8000 px.\n\nIf a single API request contains more than 20 images, a stricter per-image dimension limit applies to every image in that request. All `image` blocks in the request count toward this threshold, including images from earlier conversation turns that you resend and images nested inside `tool_result` content (for example, screenshots returned to the computer use tool). On Amazon Bedrock and Google Cloud, document blocks such as PDFs also count toward this threshold. Images exceeding the stricter limit are rejected with an `invalid_request_error` whose message references \"many-image requests\" and states the current limit in pixels. To stay under the limit on all platforms, either resize each image so that neither dimension exceeds 2000 px, or keep the request to 20 or fewer image and document blocks.\n\nThe maximum size per image is:\n\n* 10 MB (base64-encoded) when using the Claude API directly.\n* 5 MB (base64-encoded) on Amazon Bedrock and Google Cloud.\n* 10 MB on [claude.ai](https://claude.ai/).\n\n<Note>\n  Although the API supports up to 600 images per request, [request size limits](https://platform.claude.com/docs/en/api/overview#request-size-limits) ",
  "description": "Claude's vision capabilities allow it to understand and analyze images, opening up exciting possibilities for multimodal interaction.",
  "markdown_url": "https://platform.claude.com/docs/en/build-with-claude/vision.md",
  "observed_at_ms": 1792516055036,
  "outline": [
    {
      "anchor": "send-images-to-claude",
      "level": 2,
      "text": "Send images to Claude"
    },
    {
      "anchor": "base64-encoded-image-example",
      "level": 3,
      "text": "Base64-encoded image example"
    },
    {
      "anchor": "url-based-image-example",
      "level": 3,
      "text": "URL-based image example"
    },
    {
      "anchor": "files-api-image-example",
      "level": 3,
      "text": "Files API image example"
    },
    {
      "anchor": "multiple-images",
      "level": 3,
      "text": "Multiple images"
    },
    {
      "anchor": "image-limits-and-costs",
      "level": 2,
      "text": "Image limits and costs"
    },
    {
      "anchor": "request-limits",
      "level": 3,
      "text": "Request limits"
    },
    {
      "anchor": "supported-formats",
      "level": 3,
      "text": "Supported formats"
    },
    {
      "anchor": "resolution-and-token-cost",
      "level": 3,
      "text": "Resolution and token cost"
    },
    {
      "anchor": "image-quality-guidance",
      "level": 3,
      "text": "Image quality guidance"
    },
    {
      "anchor": "coordinates-and-bounding-boxes",
      "level": 2,
      "text": "Coordinates and bounding boxes"
    },
    {
      "anchor": "limitations",
      "level": 2,
      "text": "Limitations"
    },
    {
      "anchor": "faq",
      "level": 2,
      "text": "FAQ"
    },
    {
      "anchor": "next-steps",
      "level": 2,
      "text": "Next steps"
    }
  ],
  "path": "build-with-claude/vision",
  "section": {
    "anchor": "image-limits-and-costs",
    "level": 2,
    "text": "Image limits and costs"
  },
  "title": "Vision",
  "truncated": true,
  "url": "https://platform.claude.com/docs/en/build-with-claude/vision"
}
```

### List models

- Capability: `developer-docs/list_models`
- Description: The current Claude model lineup from the official models overview (`/docs/en/models/overview.md`): one record per current model with API ID, alias, description, pricing (USD per million input and output tokens, plus the raw text), context window and max output in tokens, thinking mode, default effort, knowledge and training cutoffs, retirement commitment, platform model IDs and model page; plus the legacy models still available and the table's footnotes.
- Instructions: To compare current Claude models, find a model ID for a platform, or check context window, output limit, effort default and price at a glance.
- Cost: 5 credits per call
- Capability file: [List models](https://firecrawl.dev/alexandria/agents/providers/platform-claude-com/developer-docs/list_models)

Accepted options:

Response schema example:
```json
{
  "count": 4,
  "legacy_models": [
    {
      "name": "Claude Fable 5",
      "page_url": "https://platform.claude.com/docs/en/models/fable-5/overview",
      "slug": "fable-5"
    },
    {
      "name": "Claude Opus 5",
      "page_url": "https://platform.claude.com/docs/en/models/opus-5/overview",
      "slug": "opus-5"
    },
    {
      "name": "Claude Opus 4.8",
      "page_url": "https://platform.claude.com/docs/en/models/opus-4-8/overview",
      "slug": "opus-4-8"
    },
    {
      "name": "Claude Opus 4.7",
      "page_url": "https://platform.claude.com/docs/en/models/opus-4-7/overview",
      "slug": "opus-4-7"
    },
    {
      "name": "Claude Opus 4.6",
      "page_url": "https://platform.claude.com/docs/en/models/opus-4-6/overview",
      "slug": "opus-4-6"
    },
    {
      "name": "Claude Opus 4.5",
      "page_url": "https://platform.claude.com/docs/en/models/opus-4-5/overview",
      "slug": "opus-4-5"
    },
    {
      "name": "Claude Sonnet 5",
      "page_url": "https://platform.claude.com/docs/en/models/sonnet-5/overview",
      "slug": "sonnet-5"
    },
    {
      "name": "Claude Sonnet 4.6",
      "page_url": "https://platform.claude.com/docs/en/models/sonnet-4-6/overview",
      "slug": "sonnet-4-6"
    },
    {
      "name": "Claude Haiku 4.5",
      "page_url": "https://platform.claude.com/docs/en/models/haiku-4-5/overview",
      "slug": "haiku-4-5"
    }
  ],
  "markdown_url": "https://platform.claude.com/docs/en/models/overview.md",
  "models": [
    {
      "alias": "claude-fable-5-1",
      "api_id": "claude-fable-5-1",
      "context_window": "1M tokens",
      "context_window_tokens": 1000000,
      "default_effort": "high",
      "description": "For demanding reasoning and long-horizon agentic work",
      "knowledge_cutoff": "Jun 2026",
      "latency": "Slower",
      "max_output": "128K tokens",
      "max_output_tokens": 128000,
      "name": "Claude Fable 5.1",
      "other": [],
      "page_url": "https://platform.claude.com/docs/en/models/fable-5-1/overview",
      "platform_ids": [
        {
          "model_id": "anthropic.claude-fable-5-1",
          "platform": "Amazon Bedrock"
        },
        {
          "model_id": "claude-fable-5-1",
          "platform": "Google Cloud"
        },
        {
          "model_id": "claude-fable-5-1",
          "platform": "Microsoft Foundry"
        },
        {
          "model_id": "claude-fable-5-1",
          "platform": "Claude Platform on AWS"
        }
      ],
      "pricing": {
        "input_usd_per_mtok": 10,
        "output_usd_per_mtok": 50,
        "raw": "$10 / input MTok, $50 / output MTok",
        "starting_price": false
      },
      "retirement": "Not sooner than September 1, 2027",
      "slug": "fable-5-1",
      "thinking": "Adaptive (always on)",
      "training_data_cutoff": "Jun 2026"
    },
    {
      "alias": "claude-opus-5-5",
      "api_id": "claude-opus-5-5",
      "context_window": "1M tokens",
      "context_window_tokens": 1000000,
      "default_effort": "medium",
      "description": "For long-running agentic coding and knowledge work",
      "knowledge_cutoff": "Jun 2026",
      "latency": "Moderate",
      "max_output": "128K tokens",
      "max_output_tokens": 128000,
      "name": "Claude Opus 5.5",
      "other": [],
      "page_url": "https://platform.claude.com/docs/en/models/opus-5-5/overview",
      "platform_ids": [
        {
          "model_id": "anthropic.claude-opus-5-5",
          "platform": "Amazon Bedrock"
        },
        {
          "model_id": "claude-opus-5-5",
          "platform": "Google Cloud"
        },
        {
          "model_id": "claude-opus-5-5",
          "platform": "Microsoft Foundry"
        },
        {
          "model_id": "claude-opus-5-5",
          "platform": "Claude Platform on AWS"
        }
      ],
      "pricing": {
        "input_usd_per_mtok": 4,
        "output_usd_per_mtok": 20,
        "raw": "$4 / input MTok, $20 / output MTok",
        "starting_price": false
      },
      "retirement": "Not sooner than September 22, 2027",
      "slug": "opus-5-5",
      "thinking": "Adaptive (always on)",
      "training_data_cutoff": "Jun 2026"
    },
    {
      "alias": "claude-sonnet-5-5",
      "api_id": "claude-sonnet-5-5",
      "context_window": "1M tokens",
      "context_window_tokens": 1000000,
      "default_effort": "high",
      "description": "The best combination of speed and intelligence",
      "knowledge_cutoff": "Jun 2026",
      "latency": "Fast",
      "max_output": "128K tokens",
      "max_output_tokens": 128000,
      "name": "Claude Sonnet 5.5",
      "other": [],
      "page_url": "https://platform.claude.com/docs/en/models/sonnet-5-5/overview",
      "platform_ids": [
        {
          "model_id": "anthropic.claude-sonnet-5-5",
          "platform": "Amazon Bedrock"
        },
        {
          "model_id": "claude-sonnet-5-5",
          "platform": "Google Cloud"
        },
        {
          "model_id": "claude-sonnet-5-5",
          "platform": "Microsoft Foundry"
        },
        {
          "model_id": "claude-sonnet-5-5",
          "platform": "Claude Platform on AWS"
        }
      ],
      "pricing": {
        "input_usd_per_mtok": 2,
        "output_usd_per_mtok": 10,
        "raw": "$2 / input MTok, $10 / output MTok",
        "starting_price": false
      },
      "retirement": "Not sooner than September 28, 2027",
      "slug": "sonnet-5-5",
      "thinking": "Adaptive",
      "training_data_cutoff": "Jun 2026"
    },
    {
      "alias": "claude-haiku-5-5",
      "api_id": "claude-haiku-5-5",
      "context_window": "1M tokens",
      "context_window_tokens": 1000000,
      "default_effort": "medium",
      "description": "For high-volume, latency-sensitive tasks such as classification, extraction, and routing",
      "knowledge_cutoff": "Jun 2026",
      "latency": "Fastest",
      "max_output": "128K tokens",
      "max_output_tokens": 128000,
      "name": "Claude Haiku 5.5",
      "other": [],
      "page_url": "https://platform.claude.com/docs/en/models/haiku-5-5/overview",
      "platform_ids": [
        {
          "model_id": "anthropic.claude-haiku-5-5",
          "platform": "Amazon Bedrock"
        },
        {
          "model_id": "claude-haiku-5-5",
          "platform": "Google Cloud"
        },
        {
          "model_id": "claude-haiku-5-5",
          "platform": "Microsoft Foundry"
        },
        {
          "model_id": "claude-haiku-5-5",
          "platform": "Claude Platform on AWS"
        }
      ],
      "pricing": {
        "input_usd_per_mtok": 0.1,
        "output_usd_per_mtok": 0.5,
        "raw": "From $0.10 / input MTok, From $0.50 / output MTok",
        "starting_price": true
      },
      "retirement": "Not sooner than October 7, 2027",
      "slug": "haiku-5-5",
      "thinking": "Adaptive",
      "training_data_cutoff": "Jun 2026"
    }
  ],
  "notes": [
    {
      "feature": "Comparative latency",
      "note": "Relative to the current lineup. Actual latency depends on prompt length, output length, and thinking effort."
    },
    {
      "feature": "Pricing",
      "note": "Base price per million tokens. Batch API requests are 50% off; prompt cache reads cost 10% of the base input price (2.5% on Claude Fable 5.1 and Claude Mythos 5.1, 5% on Claude Opus 5.5 and Claude Sonnet 5.5). See Pricing for cache writes, long-context, and per-platform pricing."
    },
    {
      "feature": "Claude API ID",
      "note": "Every Claude model ID is a pinned snapshot, including the dateless IDs used from the 4.6 generation on."
    },
    {
      "feature": "Thinking",
      "note": "Adaptive thinking lets the model decide how much to think, steered by effort. Extended thinking is the manual thinking.type “enabled” + budget_tokens mode on earlier models; it is deprecated on Claude Opus 4.6 and Claude Sonnet 4.6 and not accepted on later models."
    },
    {
      "feature": "Default effort",
      "note": "The effort parameter’s default on the Claude API. Set effort explicitly to use a different level."
    },
    {
      "feature": "Context window",
      "note": "1M tokens is roughly 555k words or 2.5M Unicode characters on the current tokenizer (introduced with Claude Opus 4.7); models before it fit about 750k words in 1M tokens. 200k tokens is roughly 150k words."
    },
    {
      "feature": "Max output",
      "note": "Synchronous Messages API limit. On the Message Batches API, Claude Opus 5.5, Claude Opus 5, Claude Sonnet 5.5, Claude Sonnet 5, Claude Haiku 5.5, Claude Opus 4.8, Claude Opus 4.7, Claude Opus 4.6, and Claude Sonnet 4.6 support up to 300k output tokens with the output-300k-2026-03-24 beta header."
    },
    {
      "feature": "Reliable knowledge cutoff",
      "note": "The date through which the model’s knowledge is most extensive and reliable. Training data cutoff (under Additional details) is the broader range of data used. See Anthropic’s Transparency Hub for details."
    },
    {
      "feature": "Retirement",
      "note": "Anthropic’s commitment for the platforms that Anthropic operates. Amazon Bedrock and Google Cloud set their own dates."
    },
    {
      "feature": "Claude API alias",
      "note": "For models before the 4.6 generation, the alias is a convenience pointer that resolves to the dated ID. Dateless IDs are their own pinned snapshot; the alias row repeats them."
    },
    {
      "feature": "Amazon Bedrock ID",
      "note": "The ID on Bedrock’s Messages-API endpoint (Claude Opus 4.7 and later, plus Claude Haiku 4.5); a model offered only through Bedrock’s InvokeModel integration shows that ID instead. Bedrock offers global endpoints (dynamic routing) and regional endpoints (guaranteed data routing) for Claude Sonnet 4.5 and later, and sets its own lifecycle dates."
    },
    {
      "feature": "Google Cloud ID",
      "note": "Google Cloud offers global, multi-region, and regional endpoints, and sets its own lifecycle dates."
    },
    {
      "feature": "Microsoft Foundry ID",
      "note": "Foundry deployments default to the Claude API model ID (the alias, where one exists); the deployment name is what you send. Foundry follows the Claude API lifecycle schedule."
    },
    {
      "feature": "Claude Platform on AWS ID",
      "note": "Claude Platform on AWS uses the Claude API model IDs (the dateless form where the Claude API has an alias), not Bedrock-style IDs, and follows Anthropic’s first-party model lifecycle."
    }
  ],
  "observed_at_ms": 1792516055039,
  "url": "https://platform.claude.com/docs/en/models/overview"
}
```

### Model deprecations

- Capability: `developer-docs/model_deprecations`
- Description: Model lifecycle from the official deprecations page (`/docs/en/about-claude/model-deprecations.md`): per API model ID its state (Active, Deprecated, Retired), deprecation date, retirement date or not-sooner-than commitment, recommended replacement and announcement date; every deprecation announcement row; and deprecated API parameters with their behavior (for example sampling parameters that return 400 on newer models). Filter by model substring.
- Instructions: To check whether a model is deprecated or retired, when it retires, what replaces it, or which request parameters are deprecated.
- Cost: 5 credits per call
- Capability file: [Model deprecations](https://firecrawl.dev/alexandria/agents/providers/platform-claude-com/developer-docs/model_deprecations)

Accepted options:
- `model` (string): Case-insensitive substring filter on the model name or ID, e.g. `haiku`. Example: `<model>`

Response schema example:
```json
{
  "announcements": [
    {
      "announced": "2026-02-19",
      "api_id": "claude-3-haiku-20240307",
      "recommended_replacement": "claude-haiku-4-5-20251001",
      "retirement": "April 20, 2026",
      "retirement_date": "2026-04-20",
      "title": "Claude Haiku 3 model"
    },
    {
      "announced": "2025-12-19",
      "api_id": "claude-3-5-haiku-20241022",
      "recommended_replacement": "claude-haiku-4-5-20251001",
      "retirement": "February 19, 2026",
      "retirement_date": "2026-02-19",
      "title": "Claude Haiku 3.5 model"
    }
  ],
  "count": 4,
  "markdown_url": "https://platform.claude.com/docs/en/about-claude/model-deprecations.md",
  "models": [
    {
      "announced": null,
      "api_id": "claude-haiku-5-5",
      "deprecated": null,
      "deprecated_date": null,
      "recommended_replacement": null,
      "retirement": "Not sooner than October 7, 2027",
      "retirement_date": null,
      "retirement_not_before": "2027-10-07",
      "state": "Active"
    },
    {
      "announced": null,
      "api_id": "claude-haiku-4-5-20251001",
      "deprecated": null,
      "deprecated_date": null,
      "recommended_replacement": null,
      "retirement": "Not sooner than October 15, 2026",
      "retirement_date": null,
      "retirement_not_before": "2026-10-15",
      "state": "Active"
    },
    {
      "announced": "2025-12-19",
      "api_id": "claude-3-5-haiku-20241022",
      "deprecated": "December 19, 2025",
      "deprecated_date": "2025-12-19",
      "recommended_replacement": "claude-haiku-4-5-20251001",
      "retirement": "February 19, 2026",
      "retirement_date": "2026-02-19",
      "retirement_not_before": null,
      "state": "Retired"
    },
    {
      "announced": "2026-02-19",
      "api_id": "claude-3-haiku-20240307",
      "deprecated": "February 19, 2026",
      "deprecated_date": "2026-02-19",
      "recommended_replacement": "claude-haiku-4-5-20251001",
      "retirement": "April 20, 2026",
      "retirement_date": "2026-04-20",
      "retirement_not_before": null,
      "state": "Retired"
    }
  ],
  "observed_at_ms": 1792516055051,
  "parameter_deprecations": [
    {
      "behavior": "Returns a 400 error when set to a non-default value on Claude 4.7 and later models and Claude Mythos Preview.",
      "parameters": [
        "temperature",
        "top_p",
        "top_k"
      ],
      "recommended_replacement": "Omit and use prompting to guide model behavior.",
      "status": "Deprecated (Claude 4.7 and later models)"
    }
  ],
  "url": "https://platform.claude.com/docs/en/about-claude/model-deprecations"
}
```

### Model pricing

- Capability: `developer-docs/model_pricing`
- Description: Official Claude API token prices from the pricing page (`/docs/en/about-claude/pricing.md`): per model, base input, 5-minute and 1-hour cache writes, cache hits, output, and Batch API input/output, each in USD per million tokens as a number plus the raw text, with tier notes (such as prompt-length tiers) and footnote markers; plus fast mode prices and the table footnotes. Filter by model substring.
- Instructions: To answer what a Claude model costs per million tokens, including caching and batch discounts.
- Cost: 5 credits per call
- Capability file: [Model pricing](https://firecrawl.dev/alexandria/agents/providers/platform-claude-com/developer-docs/model_pricing)

Accepted options:
- `model` (string): Case-insensitive substring filter on the model name or ID, e.g. `haiku`. Example: `<model>`

Response schema example:
```json
{
  "count": 4,
  "fast_mode": [],
  "footnotes": [
    {
      "marker": "1",
      "text": "Cache hits and refreshes on Claude Fable 5.1 and Claude Mythos 5.1 are priced at 0.025x the base input price."
    },
    {
      "marker": "2",
      "text": "Cache hits and refreshes on Claude Opus 5.5 and Claude Sonnet 5.5 are priced at 0.05x the base input price."
    },
    {
      "marker": null,
      "text": "All other models use the standard 0.1x multiplier."
    },
    {
      "marker": "3",
      "text": "The $2/$10 per million input/output token pricing for Claude Sonnet 5, announced at launch as introductory pricing through August 31, 2026, is now the standard price. The previously scheduled increase to $3/$15 per million input/output tokens on September 1, 2026 will not occur."
    }
  ],
  "markdown_url": "https://platform.claude.com/docs/en/about-claude/pricing.md",
  "models": [
    {
      "base_input": {
        "raw": "$0.10 / MTok",
        "usd_per_mtok": 0.1
      },
      "batch_input": {
        "raw": "$0.05 / MTok",
        "usd_per_mtok": 0.05
      },
      "batch_output": {
        "raw": "$0.25 / MTok",
        "usd_per_mtok": 0.25
      },
      "cache_hit": {
        "raw": "$0.01 / MTok",
        "usd_per_mtok": 0.01
      },
      "cache_write_1h": {
        "raw": "$0.20 / MTok",
        "usd_per_mtok": 0.2
      },
      "cache_write_5m": {
        "raw": "$0.125 / MTok",
        "usd_per_mtok": 0.125
      },
      "footnotes": [],
      "label": "Claude Haiku 5.5 (for prompts up to 100,000 tokens)",
      "model": "Claude Haiku 5.5",
      "note": "for prompts up to 100,000 tokens",
      "output": {
        "raw": "$0.50 / MTok",
        "usd_per_mtok": 0.5
      }
    },
    {
      "base_input": {
        "raw": "$0.50 / MTok",
        "usd_per_mtok": 0.5
      },
      "batch_input": {
        "raw": "$0.25 / MTok",
        "usd_per_mtok": 0.25
      },
      "batch_output": {
        "raw": "$1.25 / MTok",
        "usd_per_mtok": 1.25
      },
      "cache_hit": {
        "raw": "$0.05 / MTok",
        "usd_per_mtok": 0.05
      },
      "cache_write_1h": {
        "raw": "$1 / MTok",
        "usd_per_mtok": 1
      },
      "cache_write_5m": {
        "raw": "$0.625 / MTok",
        "usd_per_mtok": 0.625
      },
      "footnotes": [],
      "label": "Claude Haiku 5.5 (for prompts over 100,000 tokens)",
      "model": "Claude Haiku 5.5",
      "note": "for prompts over 100,000 tokens",
      "output": {
        "raw": "$2.50 / MTok",
        "usd_per_mtok": 2.5
      }
    },
    {
      "base_input": {
        "raw": "$1 / MTok",
        "usd_per_mtok": 1
      },
      "batch_input": {
        "raw": "$0.50 / MTok",
        "usd_per_mtok": 0.5
      },
      "batch_output": {
        "raw": "$2.50 / MTok",
        "usd_per_mtok": 2.5
      },
      "cache_hit": {
        "raw": "$0.10 / MTok",
        "usd_per_mtok": 0.1
      },
      "cache_write_1h": {
        "raw": "$2 / MTok",
        "usd_per_mtok": 2
      },
      "cache_write_5m": {
        "raw": "$1.25 / MTok",
        "usd_per_mtok": 1.25
      },
      "footnotes": [],
      "label": "Claude Haiku 4.5",
      "model": "Claude Haiku 4.5",
      "note": null,
      "output": {
        "raw": "$5 / MTok",
        "usd_per_mtok": 5
      }
    },
    {
      "base_input": {
        "raw": "$0.80 / MTok",
        "usd_per_mtok": 0.8
      },
      "batch_input": {
        "raw": "$0.40 / MTok",
        "usd_per_mtok": 0.4
      },
      "batch_output": {
        "raw": "$2 / MTok",
        "usd_per_mtok": 2
      },
      "cache_hit": {
        "raw": "$0.08 / MTok",
        "usd_per_mtok": 0.08
      },
      "cache_write_1h": {
        "raw": "$1.60 / MTok",
        "usd_per_mtok": 1.6
      },
      "cache_write_5m": {
        "raw": "$1 / MTok",
        "usd_per_mtok": 1
      },
      "footnotes": [],
      "label": "Claude Haiku 3.5 (retired, except on Google Cloud)",
      "model": "Claude Haiku 3.5",
      "note": "retired, except on Google Cloud",
      "output": {
        "raw": "$4 / MTok",
        "usd_per_mtok": 4
      }
    }
  ],
  "observed_at_ms": 1792516055046,
  "unit": "USD per million tokens",
  "url": "https://platform.claude.com/docs/en/about-claude/pricing"
}
```

### Release notes

- Capability: `developer-docs/release_notes`
- Description: Dated Claude Platform release notes (Claude API, SDKs, Console) from `/docs/en/release-notes/overview.md`, newest first: each entry's ISO date and its items as Markdown. Filter with `since` (YYYY-MM-DD, inclusive) and `q` (keeps only items containing the text); page with `limit` and `offset` (`next_offset` continues).
- Instructions: To find when an API feature, model or parameter change shipped, or what changed recently.
- Cost: 5 credits per call
- Capability file: [Release notes](https://firecrawl.dev/alexandria/agents/providers/platform-claude-com/developer-docs/release_notes)

Accepted options:
- `limit` (number): Entries per page (default 10). Example: `10`
- `offset` (number): Matched entries to skip. Example: `10`
- `q` (string): Case-insensitive text an item must contain, e.g. `budget_tokens`. Example: `budget_tokens`
- `since` (string): Only entries on or after this date. Pattern: ^[0-9]{4}-[0-9]{2}-[0-9]{2}$. Example: `<since>`

Response schema example:
```json
{
  "count": 3,
  "entries": [
    {
      "date": "2026-10-15",
      "date_raw": "October 15, 2026",
      "items": [
        "We've added `capabilities.thinking.types.between_tools` to the [Models API](https://platform.claude.com/docs/en/api/models/list). `GET /v1/models` and `GET /v1/models/{model_id}` now report whether each model accepts `thinking: {type: \"between_tools\"}`, which turns off up-front thinking. See [Using the Models API](https://platform.claude.com/docs/en/models/overview#using-the-models-api).",
        "The [Compliance API](https://platform.claude.com/docs/en/manage-claude/compliance-api) now returns the text each organization member saved as **Instructions for Claude** in claude.ai and as **Global instructions** in Cowork, in beta for Claude Enterprise organizations. Call `GET /v1/compliance/organizations/{org_uuid}/user_preferences` with your existing Compliance Access Key and the `read:compliance_user_data` scope. See [List user preferences](https://platform.claude.com/docs/en/manage-claude/compliance-org-data#list-user-preferences).",
        "[Fast mode](https://platform.claude.com/docs/en/build-with-claude/fast-mode) is out of research preview for Claude Opus 5.5, and you no longer need to request access. Requests to Claude Opus 5.5 no longer require the `fast-mode-2026-02-01` beta header, and requests that still send it keep working. Claude Opus 5.5 now has a default fast mode rate limit in the Start, Build, and Scale usage tiers. Fast mode for Claude Opus 5 and Claude Opus 4.8 remains in research preview. Keep sending the beta header with fast mode requests to those models.",
        "[Fast mode](https://platform.claude.com/docs/en/build-with-claude/fast-mode) for Claude Opus 5.5 is now available on [Claude in Microsoft Foundry](https://platform.claude.com/docs/en/build-with-claude/claude-in-microsoft-foundry) for deployments hosted on Azure."
      ]
    },
    {
      "date": "2026-10-14",
      "date_raw": "October 14, 2026",
      "items": [
        "We've launched [OpenTelemetry export](https://platform.claude.com/docs/en/manage-claude/opentelemetry) in beta, so you can trace Messages API requests and Claude Managed Agents sessions in your observability backend. The Claude SDK emits spans in Python, TypeScript, C#, Go, Java, PHP, and Ruby. To get spans from the API, [add an OpenTelemetry destination](https://platform.claude.com/docs/en/manage-claude/opentelemetry/api#add-a-destination) in the Claude Console.",
        "Claude Managed Agents [scheduled deployments](https://platform.claude.com/docs/en/managed-agents/scheduled-deployments) are now available on [Claude Platform on AWS](https://platform.claude.com/docs/en/build-with-claude/claude-platform-on-aws#claude-managed-agents)."
      ]
    },
    {
      "date": "2026-10-09",
      "date_raw": "October 9, 2026",
      "items": [
        "You can now let a Claude Managed Agents agent use [dynamic workflows](https://platform.claude.com/docs/en/managed-agents/multiagent-orchestration#dynamic-workflows), in beta with the `managed-agents-2026-04-01` beta header. For work with many pieces, such as reviewing hundreds of documents, the agent can write a workflow. A workflow is a program that runs many agents in phases and combines their results. The server runs it in the background as a workflow run. To [turn dynamic workflows on](https://platform.claude.com/docs/en/managed-agents/multiagent-orchestration#turn-on-dynamic-workflows), set the agent's `multiagent` field to `{\"type\": \"multiagent_20261001\", \"workflows\": {\"type\": \"enabled\"}}`. To guide the agent, tell it in its system prompt when to start a run. Follow each run through `workflow_run.*` events on the session's event stream. See [Workflow runs](https://platform.claude.com/docs/en/managed-agents/workflow-runs)."
      ]
    }
  ],
  "markdown_url": "https://platform.claude.com/docs/en/release-notes/overview.md",
  "matched_entries": 154,
  "newest_date": "2026-10-15",
  "next_offset": 3,
  "observed_at_ms": 1792516055054,
  "query": {
    "limit": 3,
    "offset": 0,
    "q": null,
    "since": null
  },
  "total_entries": 154,
  "url": "https://platform.claude.com/docs/en/release-notes/overview"
}
```

### Search docs

- Capability: `developer-docs/search_docs`
- Description: Search the official Claude Developer Platform documentation (guides and the API reference) through the docs' own search, the `search_docs` tool of Anthropic's public docs MCP server. Returns ranked results with title, breadcrumb, kind, URL, docs path, description, endpoint (for API methods) and a snippet, in the vendor's ranking order.
- Instructions: To find the official page for a Claude API topic (thinking budgets, sampling parameters, vision, PDF input, tool use, prompt caching, errors) before reading it with get_page.
- Cost: 5 credits per call
- Capability file: [Search docs](https://firecrawl.dev/alexandria/agents/providers/platform-claude-com/developer-docs/search_docs)

Accepted options:
- `kind` (string): Only guide pages, API methods or API resources. Example: `prose_page`
- `limit` (number): Maximum results (default 10). Example: `10`
- `query` (string, required): Keywords or phrase, e.g. `extended thinking budget_tokens`. Example: `extended thinking budget_tokens`
- `sdk_language` (string): Search this SDK's API reference with the guides, e.g. `python`, `typescript`; default is the HTTP reference. Pattern: ^[A-Za-z#+-]+$. Example: `<sdk_language>`

Response schema example:
```json
{
  "count": 5,
  "kind": null,
  "limit": 5,
  "observed_at_ms": 1792516055035,
  "query": "extended thinking budget_tokens",
  "results": [
    {
      "breadcrumb": [
        "Messages",
        "Thinking",
        "Extended thinking"
      ],
      "description": "Configure manual extended thinking with a fixed budget_tokens budget on Claude models that support it, and migrate to adaptive thinking.",
      "endpoint": null,
      "kind": "prose_page",
      "path": "build-with-claude/extended-thinking",
      "rank": 1,
      "snippet": "To learn how zero data retention (ZDR) applies to this feature, see API and data retention . Extended thinking ( thinking.type: \"enabled\" with budget_tokens ) is deprecated on the Claude 4.6 models (r...",
      "title": "Extended thinking",
      "url": "https://platform.claude.com/docs/en/build-with-claude/extended-thinking"
    },
    {
      "breadcrumb": [
        "Messages",
        "Thinking",
        "Thinking"
      ],
      "description": "Understand how Claude's thinking works: turn it on, read thinking output, steer thinking depth with effort, and use thinking with tools, caching, and streaming.",
      "endpoint": null,
      "kind": "prose_page",
      "path": "build-with-claude/thinking",
      "rank": 2,
      "snippet": "To learn how zero data retention (ZDR) applies to this feature, see API and data retention . A model that answers in a single pass has to get everything right on the first try: no scratch work, no che...",
      "title": "Thinking",
      "url": "https://platform.claude.com/docs/en/build-with-claude/thinking"
    },
    {
      "breadcrumb": [
        "Messages",
        "Thinking",
        "Preserved thinking"
      ],
      "description": "Preserved thinking lets a model use a thinking block from an earlier turn only if that model or one of a fixed set of other models produced it and nothing befor…",
      "endpoint": null,
      "kind": "prose_page",
      "path": "build-with-claude/preserved-thinking",
      "rank": 3,
      "snippet": "Preserved thinking is a property of newer Claude models that guards against distillation. It determines whether the model can use a thinking block that you send back from an earlier turn. Starting wit...",
      "title": "Preserved thinking",
      "url": "https://platform.claude.com/docs/en/build-with-claude/preserved-thinking"
    },
    {
      "breadcrumb": [
        "Messages",
        "Model capabilities",
        "Task budgets"
      ],
      "description": "Give Claude an advisory token budget for the full agentic loop to help the model self-regulate on long agentic tasks.",
      "endpoint": null,
      "kind": "prose_page",
      "path": "build-with-claude/task-budgets",
      "rank": 4,
      "snippet": "Task budgets let you tell Claude how many tokens it has for a full agentic loop, including thinking, tool calls, tool results, and output. The model sees a running countdown and uses it to prioritize ...",
      "title": "Task budgets",
      "url": "https://platform.claude.com/docs/en/build-with-claude/task-budgets"
    },
    {
      "breadcrumb": [
        "Messages",
        "Thinking",
        "Troubleshooting thinking"
      ],
      "description": "Diagnose and fix the most common thinking failures: configuration 400 errors, empty or missing thinking blocks, max_tokens stops, and cache misses.",
      "endpoint": null,
      "kind": "prose_page",
      "path": "build-with-claude/thinking-troubleshooting",
      "rank": 5,
      "snippet": "To learn how zero data retention (ZDR) applies to this feature, see API and data retention . This page covers the most common failures when configuring thinking or round-tripping thinking blocks (send...",
      "title": "Troubleshooting thinking",
      "url": "https://platform.claude.com/docs/en/build-with-claude/thinking-troubleshooting"
    }
  ],
  "sdk_language": null,
  "source_url": "https://platform.claude.com/docs/mcp"
}
```