@ai-sdk/anthropic 4.0.62 → 4.0.65

package/docs/05-anthropic.mdx

@@ -64,18 +64,18 @@ You can use the following optional settings to customize the Anthropic provider
 ## Language Models
 
 You can create models that call the [Anthropic Messages API](https://docs.anthropic.com/claude/reference/messages_post) using the provider instance.
-The first argument is the model id, e.g. `claude-3-haiku-20240307`.
+The first argument is the model id, e.g. `claude-haiku-4-5`.
 Some models have multi-modal capabilities.
 
 ```ts
-const model = anthropic('claude-3-haiku-20240307');
+const model = anthropic('claude-haiku-4-5');
 ```
 
 You can also use the following aliases for model creation:
 
-- `anthropic.languageModel('claude-3-haiku-20240307')` - Creates a language model
-- `anthropic.chat('claude-3-haiku-20240307')` - Alias for `languageModel`
-- `anthropic.messages('claude-3-haiku-20240307')` - Alias for `languageModel`
+- `anthropic.languageModel('claude-haiku-4-5')` - Creates a language model
+- `anthropic.chat('claude-haiku-4-5')` - Alias for `languageModel`
+- `anthropic.messages('claude-haiku-4-5')` - Alias for `languageModel`
 
 You can use Anthropic language models to generate text with the `generateText` function:
 
@@ -84,7 +84,7 @@ import { anthropic } from '@ai-sdk/anthropic';
 import { generateText } from 'ai';
 
 const { text } = await generateText({
-  model: anthropic('claude-3-haiku-20240307'),
+  model: anthropic('claude-haiku-4-5'),
   prompt: 'Write a vegetarian lasagna recipe for 4 people.',
 });
 ```
@@ -179,7 +179,7 @@ import { streamText, tool } from 'ai';
 import { z } from 'zod';
 
 const result = streamText({
-  model: anthropic('claude-sonnet-4-20250514'),
+  model: anthropic('claude-sonnet-5'),
   tools: {
     writeFile: tool({
       description: 'Write content to a file',
@@ -223,7 +223,7 @@ import { anthropic, AnthropicLanguageModelOptions } from '@ai-sdk/anthropic';
 import { generateText } from 'ai';
 
 const { text, usage } = await generateText({
-  model: anthropic('claude-opus-4-20250514'),
+  model: anthropic('claude-opus-5-5'),
   prompt: 'How many people will live in the world in 2040?',
   providerOptions: {
     anthropic: {
@@ -268,7 +268,7 @@ import { anthropic, AnthropicLanguageModelOptions } from '@ai-sdk/anthropic';
 import { generateText } from 'ai';
 
 const { text } = await generateText({
-  model: anthropic('claude-opus-4-8'),
+  model: anthropic('claude-opus-5-5'),
   prompt: 'Research the pros and cons of Rust vs Go for building CLI tools.',
   providerOptions: {
     anthropic: {
@@ -306,7 +306,7 @@ import { anthropic, AnthropicLanguageModelOptions } from '@ai-sdk/anthropic';
 import { generateText } from 'ai';
 
 const { text } = await generateText({
-  model: anthropic('claude-opus-4-6'),
+  model: anthropic('claude-opus-5-5'),
   prompt: 'Summarize the key points of this document.',
   providerOptions: {
     anthropic: {
@@ -351,7 +351,7 @@ import { anthropic, AnthropicLanguageModelOptions } from '@ai-sdk/anthropic';
 import { generateText } from 'ai';
 
 const { text } = await generateText({
-  model: anthropic('claude-fable-5'),
+  model: anthropic('claude-fable-5-1'),
   prompt: 'Explain the history of cryptography.',
   providerOptions: {
     anthropic: {
@@ -368,11 +368,11 @@ import { anthropic, AnthropicLanguageModelOptions } from '@ai-sdk/anthropic';
 import { generateText } from 'ai';
 
 const { text } = await generateText({
-  model: anthropic('claude-fable-5'),
+  model: anthropic('claude-fable-5-1'),
   prompt: 'Explain the history of cryptography.',
   providerOptions: {
     anthropic: {
-      fallbacks: [{ model: 'claude-fable-5' }],
+      fallbacks: [{ model: 'claude-fable-5-1' }],
     } satisfies AnthropicLanguageModelOptions,
   },
 });
@@ -405,7 +405,7 @@ import { generateText, tool } from 'ai';
 import { z } from 'zod';
 
 const result = await generateText({
-  model: anthropic('claude-sonnet-4-5'),
+  model: anthropic('claude-sonnet-5'),
   prompt: 'Use the bash tool to run: echo hello',
   tools: { bash: tool({ inputSchema: z.object({ command: z.string() }) }) },
   providerOptions: {
@@ -715,7 +715,89 @@ const result = await generateText({
 });
 ```
 
-#### Compaction
+#### On-Demand Compaction
+
+Use the `compaction` provider option to ask Anthropic to summarize the supplied
+conversation immediately. The provider automatically adds the
+`compact-2026-09-04` beta header.
+
+```ts highlight="10-15,21-22"
+import { anthropic, AnthropicLanguageModelOptions } from '@ai-sdk/anthropic';
+import { generateText, ModelMessage } from 'ai';
+
+const history: ModelMessage[] = [
+  { role: 'user', content: 'Help me design a recipe application.' },
+  { role: 'assistant', content: 'Start with Recipe and Ingredient.' },
+  { role: 'user', content: 'Recipe also needs a title and servings.' },
+];
+
+const summary = await generateText({
+  model: anthropic('claude-opus-5'),
+  messages: history,
+  maxOutputTokens: 4096,
+  providerOptions: {
+    anthropic: {
+      compaction: {
+        type: 'summarize',
+        instructions: 'Preserve all agreed entities and fields.',
+      },
+    } satisfies AnthropicLanguageModelOptions,
+  },
+});
+
+if (summary.rawFinishReason === 'compaction') {
+  const [compactionMessage] = summary.finalStep.response.messages;
+
+  const continuation = await generateText({
+    model: anthropic('claude-opus-5'),
+    messages: [
+      compactionMessage,
+      { role: 'user', content: 'Which fields should be required?' },
+    ],
+  });
+
+  console.log(continuation.text);
+}
+```
+
+Anthropic returns the summary as a text part whose
+`providerMetadata.anthropic.type` is `compaction`. Its opaque signature is
+available as `providerMetadata.anthropic.signature`. Response messages copy
+that metadata to `providerOptions`, so passing
+`summary.finalStep.response.messages` back to the AI SDK preserves the signed
+block and automatically adds the beta header on the continuation request.
+
+Check `rawFinishReason` before reading the response block. Anthropic can return
+a successful HTTP response without a summary when summarization ends for
+another reason, such as `max_tokens`, `model_context_window_exceeded`,
+`tool_use`, `refusal`, or `end_turn`. In that case, keep the existing history
+and retry later; the unsuccessful summarization is still included in
+`providerMetadata.anthropic.iterations`.
+
+Send exactly one signed compaction block on continuation requests. It must be
+the first content block in `messages` and must be replayed without changing its
+text or signature. Remove the messages it summarizes before continuing. Keep
+the same system prompt and tool definitions on the compaction request that you
+use for the conversation, especially when preserving later thinking blocks.
+
+Before requesting compaction, resolve any unfinished tool call from the last
+assistant turn. Do not combine `compaction` with `contextManagement`,
+`stopSequences`, structured output, a required or named `toolChoice`, or a
+`taskBudget.remaining` value. Also omit `taskBudget.remaining` from later
+requests that carry the signed block.
+
+On-demand compaction usage is reported in
+`providerMetadata.anthropic.iterations`. Sum the iteration usage when
+calculating the tokens consumed by the summarization request.
+
+On-demand compaction can also be initiated in a Message Batch. Configure the
+`compaction` provider option on the individual batch request; the provider
+automatically adds `compact-2026-09-04` to the batch beta header.
+
+For more details, see
+[Anthropic's on-demand compaction documentation](https://platform.claude.com/docs/en/build-with-claude/compaction-on-demand).
+
+#### Threshold Compaction
 
 The `compact_20260112` edit type automatically summarizes earlier conversation context when token limits are reached. This is useful for long-running conversations where you want to preserve the essence of earlier exchanges while staying within token limits.
 
@@ -724,7 +806,7 @@ import { anthropic, AnthropicLanguageModelOptions } from '@ai-sdk/anthropic';
 import { streamText } from 'ai';
 
 const result = streamText({
-  model: anthropic('claude-opus-4-6'),
+  model: anthropic('claude-opus-5-5'),
   messages: conversationHistory,
   providerOptions: {
     anthropic: {
@@ -848,7 +930,7 @@ import { generateText } from 'ai';
 const errorMessage = '... long error message ...';
 
 const result = await generateText({
-  model: anthropic('claude-sonnet-4-5'),
+  model: anthropic('claude-sonnet-5'),
   messages: [
     {
       role: 'user',
@@ -882,7 +964,7 @@ You can also use cache control on system messages by providing multiple system m
 
 ```ts highlight="3,7-9"
 const result = await generateText({
-  model: anthropic('claude-sonnet-4-5'),
+  model: anthropic('claude-sonnet-5'),
   messages: [
     {
       role: 'system',
@@ -1219,7 +1301,7 @@ const webSearchTool = anthropic.tools.webSearch_20260318({
 });
 
 const result = await generateText({
-  model: anthropic('claude-opus-5'),
+  model: anthropic('claude-opus-5-5'),
   prompt: 'What are the latest developments in AI?',
   tools: {
     web_search: webSearchTool,
@@ -1278,7 +1360,7 @@ const webSearchTool = anthropic.tools.webSearch_20260318({
 });
 
 const result = await generateText({
-  model: anthropic('claude-opus-5'),
+  model: anthropic('claude-opus-5-5'),
   prompt: 'Find local news about technology',
   tools: {
     web_search: webSearchTool,
@@ -1304,7 +1386,7 @@ import { anthropic } from '@ai-sdk/anthropic';
 import { generateText } from 'ai';
 
 const result = await generateText({
-  model: anthropic('claude-opus-5'),
+  model: anthropic('claude-opus-5-5'),
   prompt:
     'What is this page about? https://en.wikipedia.org/wiki/Maglemosian_culture',
   tools: {
@@ -1396,7 +1478,7 @@ import { generateText, tool } from 'ai';
 import { z } from 'zod';
 
 const result = await generateText({
-  model: anthropic('claude-sonnet-4-5'),
+  model: anthropic('claude-sonnet-5'),
   prompt: 'What is the weather in San Francisco?',
   tools: {
     toolSearch: anthropic.tools.toolSearchBm25_20251119(),
@@ -1426,7 +1508,7 @@ For more precise tool matching, you can use the regex variant:
 
 ```ts
 const result = await generateText({
-  model: anthropic('claude-sonnet-4-5'),
+  model: anthropic('claude-sonnet-5'),
   prompt: 'Get the weather data',
   tools: {
     toolSearch: anthropic.tools.toolSearchRegex_20251119(),
@@ -1447,7 +1529,7 @@ import { generateText, tool } from 'ai';
 import { z } from 'zod';
 
 const result = await generateText({
-  model: anthropic('claude-sonnet-4-5'),
+  model: anthropic('claude-sonnet-5'),
   prompt: 'What is the weather in San Francisco?',
   tools: {
     // Custom search tool
@@ -1490,13 +1572,16 @@ This sends `tool_reference` blocks to Anthropic, which loads the corresponding d
 
 ### Mid-Conversation System Controls
 
-With `claude-opus-5-5` and `claude-fable-5-1`, a mid-conversation system message
-can be cleared before the next user message with `clearAt: 'next_user_message'`,
-or set the effort for the next turn without invalidating the prompt cache. The provider adds the required
+With `claude-opus-5-5` and `claude-fable-5-1`, use an empty system message to
+change effort from the next user turn. A separate system message containing text
+can provide a turn-scoped reminder with `clearAt: 'next_user_message'`: its text
+applies until the next user message. Keep the reminder after a user turn and
+before an assistant turn, or at the end of the messages array.
+The provider adds the required
 `mid-conversation-system-clear-at-2026-08-21` and
 `mid-conversation-output-config-2026-07-01` beta headers automatically.
 
-```ts highlight="17-25"
+```ts highlight="16-24,29-37"
 import { anthropic } from '@ai-sdk/anthropic';
 import { generateText } from 'ai';
 
@@ -1514,10 +1599,9 @@ const result = await generateText({
     },
     {
       role: 'system',
-      content: 'For the next turn, verify every compatibility claim.',
+      content: '',
       providerOptions: {
         anthropic: {
-          clearAt: 'next_user_message',
           effort: 'high',
         },
       },
@@ -1526,19 +1610,39 @@ const result = await generateText({
       role: 'user',
       content: 'Complete the plan.',
     },
+    {
+      role: 'system',
+      content: 'For this turn, verify every compatibility claim.',
+      providerOptions: {
+        anthropic: {
+          clearAt: 'next_user_message',
+        },
+      },
+    },
   ],
 });
 ```
 
 An effort-only system message can use an empty content array at the API level;
 in an AI SDK prompt, set `content: ''`. The provider serializes it as
-`content: []`.
+`content: []`. These effort-only messages can appear first or later in the
+conversation, including consecutively. The provider preserves their order. The new
+level applies from the next user turn until a later message changes it.
+Initial instruction text is sent in the top-level `system` field.
+
+Consecutive inline system messages are evaluated as a group. If an effort-only
+message is adjacent to a system message containing text, the whole group must
+follow Anthropic's [content placement rules](https://platform.claude.com/docs/en/build-with-claude/mid-conversation-system-messages#limitations).
+
+Turn-scoped messages with `clearAt` support text only. Put effort updates in a
+separate message, as in the example above.
 
 <Note>
-  `clearAt` and per-message `effort` apply only to mid-conversation system
-  messages. When used on the initial system message, the provider ignores them
-  and emits a warning. Use the top-level `effort` provider option to configure
-  effort for the full request.
+  `clearAt` applies only to mid-conversation system messages. On initial system
+  messages, the provider ignores `clearAt` and effort attached to non-empty
+  content and emits a warning. Use an empty system message for an initial
+  per-message effort setting, or the top-level `effort` provider option to
+  configure effort for the full request.
 </Note>
 
 ### Mid-Conversation Tool Changes
@@ -1607,7 +1711,7 @@ import { anthropic, AnthropicLanguageModelOptions } from '@ai-sdk/anthropic';
 import { generateText } from 'ai';
 
 const result = await generateText({
-  model: anthropic('claude-sonnet-4-5'),
+  model: anthropic('claude-sonnet-5'),
   prompt: `Call the echo tool with "hello world". what does it respond with back?`,
   providerOptions: {
     anthropic: {
@@ -1679,7 +1783,7 @@ Web search errors throw exceptions that you can catch:
 ```ts
 try {
   const result = await generateText({
-    model: anthropic('claude-opus-4-20250514'),
+    model: anthropic('claude-opus-5-5'),
     prompt: 'Search for something',
     tools: {
       web_search: webSearchTool,
@@ -1698,7 +1802,7 @@ Web search errors are delivered as error parts in the stream:
 
 ```ts
 const result = await streamText({
-  model: anthropic('claude-opus-4-20250514'),
+  model: anthropic('claude-opus-5-5'),
   prompt: 'Search for something',
   tools: {
     web_search: webSearchTool,
@@ -1793,7 +1897,7 @@ Code execution errors are delivered as tool result parts in the response:
 
 ```ts
 const result = await generateText({
-  model: anthropic('claude-opus-4-20250514'),
+  model: anthropic('claude-opus-5-5'),
   prompt: 'Execute some Python script',
   tools: {
     code_execution: codeExecutionTool,
@@ -1818,7 +1922,7 @@ Code execution errors are delivered as error parts in the stream:
 
 ```ts
 const result = await streamText({
-  model: anthropic('claude-opus-4-20250514'),
+  model: anthropic('claude-opus-5-5'),
   prompt: 'Execute some Python script',
   tools: {
     code_execution: codeExecutionTool,
@@ -1978,7 +2082,7 @@ Option 1: URL-based PDF document
 
 ```ts
 const result = await generateText({
-  model: anthropic('claude-sonnet-4-5'),
+  model: anthropic('claude-sonnet-5'),
   messages: [
     {
       role: 'user',
@@ -2004,7 +2108,7 @@ Option 2: Base64-encoded PDF document
 
 ```ts
 const result = await generateText({
-  model: anthropic('claude-sonnet-4-5'),
+  model: anthropic('claude-sonnet-5'),
   messages: [
     {
       role: 'user',

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@ai-sdk/anthropic",
-  "version": "4.0.62",
+  "version": "4.0.65",
   "type": "module",
   "license": "Apache-2.0",
   "sideEffects": false,
@@ -36,10 +36,10 @@
   },
   "dependencies": {
     "@ai-sdk/provider": "4.0.18",
-    "@ai-sdk/provider-utils": "5.0.47"
+    "@ai-sdk/provider-utils": "5.0.49"
   },
   "devDependencies": {
-    "@ai-sdk/test-server": "2.0.1",
+    "@ai-sdk/test-server": "2.0.2",
     "@types/node": "22.19.19",
     "@vercel/ai-tsconfig": "0.0.0",
     "tsup": "^8.5.1",

package/src/anthropic-api.ts

@@ -78,6 +78,7 @@ export interface AnthropicAssistantMessage {
 export interface AnthropicCompactionContent {
   type: 'compaction';
   content: string;
+  signature?: string;
   cache_control?: AnthropicCacheControl;
 }
 
@@ -805,6 +806,7 @@ export const anthropicResponseSchema = lazySchema(() =>
           z.object({
             type: z.literal('compaction'),
             content: z.string().nullish(),
+            signature: z.string().nullish(),
           }),
           z.object({
             type: z.literal('tool_use'),
@@ -1187,6 +1189,7 @@ export const anthropicChunkSchema = lazySchema(() =>
           z.object({
             type: z.literal('compaction'),
             content: z.string().nullish(),
+            signature: z.string().nullish(),
           }),
           z.object({
             type: z.literal('server_tool_use'),

package/src/anthropic-batch.ts

@@ -806,7 +806,12 @@ function convertAnthropicBatchResponse(
         content.push({
           type: 'text',
           text: part.content,
-          providerMetadata: { anthropic: { type: 'compaction' } },
+          providerMetadata: {
+            anthropic: {
+              type: 'compaction',
+              ...(part.signature != null && { signature: part.signature }),
+            },
+          },
         });
         break;
       case 'tool_use':

package/src/anthropic-language-model-options.ts

@@ -82,9 +82,10 @@ export const anthropicSystemMessageProviderOptions = z.object({
   clearAt: z.literal('next_user_message').optional(),
 
   /**
-   * Sets the model effort for the turn that follows this mid-conversation
-   * system message. The required `mid-conversation-output-config-2026-07-01`
-   * beta is added automatically.
+   * Sets the model effort from the next user turn until a later message
+   * changes it. An effort-only system message with empty content can appear
+   * first. The required `mid-conversation-output-config-2026-07-01` beta is
+   * added automatically.
    */
   effort: z.enum(['low', 'medium', 'high', 'xhigh', 'max']).optional(),
 
@@ -384,6 +385,19 @@ export const anthropicLanguageModelOptions = z.object({
     )
     .optional(),
 
+  /**
+   * Requests an on-demand summary of the supplied conversation.
+   *
+   * The required `compact-2026-09-04` beta is added automatically.
+   * Cannot be combined with `contextManagement`.
+   */
+  compaction: z
+    .object({
+      type: z.literal('summarize'),
+      instructions: z.string().optional(),
+    })
+    .optional(),
+
   contextManagement: z
     .object({
       edits: z.array(

package/src/anthropic-language-model.ts

@@ -1,5 +1,6 @@
 import {
   APICallError,
+  InvalidArgumentError,
   InvalidResponseDataError,
   type JSONObject,
   type LanguageModelV4,
@@ -508,6 +509,15 @@ export class AnthropicLanguageModel implements LanguageModelV4 {
     }
 
     const contextManagement = anthropicOptions?.contextManagement;
+    const compaction = anthropicOptions?.compaction;
+
+    if (contextManagement != null && compaction != null) {
+      throw new InvalidArgumentError({
+        argument: 'providerOptions',
+        message:
+          'Anthropic provider options `compaction` and `contextManagement` cannot be used together.',
+      });
+    }
 
     // Create a shared cache control validator to track breakpoints across tools and messages
     const cacheControlValidator = new CacheControlValidator();
@@ -779,6 +789,15 @@ export class AnthropicLanguageModel implements LanguageModelV4 {
           })),
         }),
 
+      ...(compaction && {
+        compaction: {
+          type: compaction.type,
+          ...(compaction.instructions !== undefined && {
+            instructions: compaction.instructions,
+          }),
+        },
+      }),
+
       ...(contextManagement && {
         context_management: {
           edits: contextManagement.edits
@@ -925,6 +944,10 @@ export class AnthropicLanguageModel implements LanguageModelV4 {
       betas.add('dangerous-tool-use-2026-09-03');
     }
 
+    if (compaction) {
+      betas.add('compact-2026-09-04');
+    }
+
     if (contextManagement) {
       betas.add('context-management-2025-06-27');
 
@@ -1271,6 +1294,9 @@ export class AnthropicLanguageModel implements LanguageModelV4 {
             providerMetadata: {
               anthropic: {
                 type: 'compaction',
+                ...(part.signature != null && {
+                  signature: part.signature,
+                }),
               },
             },
           });
@@ -1971,9 +1997,27 @@ export class AnthropicLanguageModel implements LanguageModelV4 {
                     providerMetadata: {
                       anthropic: {
                         type: 'compaction',
+                        ...(part.signature != null && {
+                          signature: part.signature,
+                        }),
                       },
                     },
                   });
+
+                  // On-demand compaction blocks arrive complete in
+                  // content_block_start, without compaction_delta events.
+                  if (
+                    part.signature != null &&
+                    part.content != null &&
+                    part.content !== ''
+                  ) {
+                    controller.enqueue({
+                      type: 'text-delta',
+                      id: String(value.index),
+                      delta: part.content,
+                    });
+                  }
+
                   return;
                 }
 

package/src/convert-to-anthropic-prompt.ts

@@ -204,11 +204,6 @@ export async function convertToAnthropicPrompt({
           });
         }
 
-        // The first block becomes the top-level system prompt. Later system
-        // blocks are sent as inline system messages — always when they carry
-        // tool changes (which are only valid mid-conversation), and otherwise
-        // only when a top-level system prompt already exists (preserving the
-        // existing hoisting behavior for plain text).
         const toolChangeCount = convertedMessages.reduce(
           (count, message) => count + message.toolChangeCount,
           0,
@@ -230,12 +225,26 @@ export async function convertToAnthropicPrompt({
             });
           }
 
+          // Initial instruction text goes in the top-level system field.
+          // Effort-only messages stay in the messages array.
           for (const message of convertedMessages) {
-            if (message.clearAt != null || message.effort != null) {
+            if (
+              message.content.length === 0 &&
+              message.clearAt == null &&
+              message.effort != null
+            ) {
+              messages.push({
+                role: 'system',
+                content: [],
+                output_config: { effort: message.effort },
+              });
+              betas.add('mid-conversation-output-config-2026-07-01');
+            } else if (message.clearAt != null || message.effort != null) {
               warnings.push({
                 type: 'other',
                 message:
-                  'clearAt and effort on the initial system message are not supported by Anthropic. ' +
+                  'clearAt and effort on this initial system message are not supported by Anthropic. ' +
+                  'Use a separate effort-only system message with empty content to set effort. ' +
                   'These options have been ignored.',
               });
             }
@@ -711,7 +720,11 @@ export async function convertToAnthropicPrompt({
               case 'text': {
                 // Check if this is a compaction block (via providerMetadata)
                 const textMetadata = part.providerOptions?.anthropic as
-                  | { type?: string; citations?: Citation[] }
+                  | {
+                      type?: string;
+                      citations?: Citation[];
+                      signature?: string;
+                    }
                   | undefined;
 
                 if (textMetadata?.type === 'compaction') {
@@ -719,9 +732,16 @@ export async function convertToAnthropicPrompt({
                     break;
                   }
 
+                  if (typeof textMetadata.signature === 'string') {
+                    betas.add('compact-2026-09-04');
+                  }
+
                   anthropicContent.push({
                     type: 'compaction',
                     content: part.text,
+                    ...(typeof textMetadata.signature === 'string' && {
+                      signature: textMetadata.signature,
+                    }),
                     cache_control: cacheControl,
                   });
                 } else {