chore (ai/core): move streamText toolCallStreaming option to stable (#4493)

Author: lgrammel

.changeset/lucky-garlics-jog.md

@@ -0,0 +1,5 @@
+---
+'ai': patch
+---
+
+chore (ai/core): move streamText toolCallStreaming option to stable

content/docs/03-ai-sdk-core/05-generating-text.mdx

@@ -104,8 +104,8 @@ It receives the following chunk types:
 - `reasoning`
 - `tool-call`
 - `tool-result`
-- `tool-call-streaming-start` (when `experimental_toolCallStreaming` is enabled)
-- `tool-call-delta` (when `experimental_toolCallStreaming` is enabled)
+- `tool-call-streaming-start` (when `toolCallStreaming` is enabled)
+- `tool-call-delta` (when `toolCallStreaming` is enabled)
 
 ```tsx highlight="6-11"
 import { streamText } from 'ai';

content/docs/04-ai-sdk-ui/03-chatbot-tool-usage.mdx

@@ -192,17 +192,15 @@ export default function Chat() {
 
 ## Tool call streaming
 
-<Note type="warning">This feature is experimental.</Note>
-
 You can stream tool calls while they are being generated by enabling the
-`experimental_toolCallStreaming` option in `streamText`.
+`toolCallStreaming` option in `streamText`.
 
 ```tsx filename='app/api/chat/route.ts' highlight="5"
 export async function POST(req: Request) {
   // ...
 
   const result = streamText({
-    experimental_toolCallStreaming: true,
+    toolCallStreaming: true,
     // ...
   });
 

content/docs/07-reference/01-ai-sdk-core/02-stream-text.mdx

@@ -470,7 +470,7 @@ To see `streamText` in action, check out [these examples](#examples).
       ],
     },
     {
-      name: 'experimental_toolCallStreaming',
+      name: 'toolCallStreaming',
       type: 'boolean',
       isOptional: true,
       description:

examples/ai-core/src/stream-text/openai-on-chunk-tool-call-streaming.ts

@@ -13,7 +13,7 @@ async function main() {
         parameters: z.object({ city: z.string() }),
       },
     },
-    experimental_toolCallStreaming: true,
+    toolCallStreaming: true,
     onChunk(chunk) {
       console.log('onChunk', chunk);
     },

examples/next-openai/app/api/use-chat-streaming-tool-calls/route.ts

@@ -11,7 +11,7 @@ export async function POST(req: Request) {
   const result = streamText({
     model: openai('gpt-4-turbo'),
     messages,
-    experimental_toolCallStreaming: true,
+    toolCallStreaming: true,
     system:
       'You are a helpful assistant that answers questions about the weather in a given city.' +
       'You use the showWeatherInformation tool to show the weather information to the user instead of talking about it.',

examples/next-openai/app/api/use-chat-tools/route.ts

@@ -11,7 +11,7 @@ export async function POST(req: Request) {
   const result = streamText({
     model: openai('gpt-4o'),
     messages,
-    experimental_toolCallStreaming: true,
+    toolCallStreaming: true,
     maxSteps: 5, // multi-steps for server-side tools
     tools: {
       // server-side tool with execute function:

examples/nuxt-openai/server/api/use-chat-tools.ts

@@ -13,7 +13,7 @@ export default defineLazyEventHandler(async () => {
     const result = streamText({
       model: openai('gpt-4-turbo'),
       messages,
-      experimental_toolCallStreaming: true,
+      toolCallStreaming: true,
       maxSteps: 5, // multi-steps for server-side tools
       tools: {
         // server-side tool with execute function:

packages/ai/core/generate-text/stream-text.test.ts

@@ -526,7 +526,7 @@ describe('streamText', () => {
 
     it('should send tool call deltas when toolCallStreaming is enabled', async () => {
       const result = streamText({
-        experimental_toolCallStreaming: true,
+        toolCallStreaming: true,
         model: createTestModel({
           stream: convertArrayToReadableStream([
             {
@@ -1108,7 +1108,7 @@ describe('streamText', () => {
           },
         },
         prompt: 'test-input',
-        experimental_toolCallStreaming: true,
+        toolCallStreaming: true,
         experimental_generateMessageId: mockId({ prefix: 'msg' }),
       });
 
@@ -1826,7 +1826,7 @@ describe('streamText', () => {
           },
         },
         prompt: 'test-input',
-        experimental_toolCallStreaming: true,
+        toolCallStreaming: true,
         onChunk(event) {
           result.push(event.chunk);
         },
@@ -3800,7 +3800,7 @@ describe('streamText', () => {
             },
           },
           prompt: 'test-input',
-          experimental_toolCallStreaming: true,
+          toolCallStreaming: true,
           onChunk(event) {
             result.push(event.chunk);
           },

packages/ai/core/generate-text/stream-text.ts

@@ -145,7 +145,8 @@ export function streamText<
   experimental_continueSteps: continueSteps = false,
   experimental_telemetry: telemetry,
   experimental_providerMetadata: providerMetadata,
-  experimental_toolCallStreaming: toolCallStreaming = false,
+  experimental_toolCallStreaming = false,
+  toolCallStreaming = experimental_toolCallStreaming,
   experimental_activeTools: activeTools,
   experimental_repairToolCall: repairToolCall,
   experimental_transform: transform,
@@ -227,6 +228,11 @@ A function that attempts to repair a tool call that failed to parse.
     /**
 Enable streaming of tool call deltas as they are generated. Disabled by default.
      */
+    toolCallStreaming?: boolean;
+
+    /**
+@deprecated Use `toolCallStreaming` instead.
+     */
     experimental_toolCallStreaming?: boolean;
 
     /**