@ai-sdk/openai 4.0.3 → 4.0.5

package/CHANGELOG.md

@@ -1,5 +1,20 @@
 # @ai-sdk/openai
 
+## 4.0.5
+
+### Patch Changes
+
+- Updated dependencies [8c616f0]
+  - @ai-sdk/provider-utils@5.0.3
+
+## 4.0.4
+
+### Patch Changes
+
+- Updated dependencies [0274f34]
+  - @ai-sdk/provider@4.0.1
+  - @ai-sdk/provider-utils@5.0.2
+
 ## 4.0.3
 
 ### Patch Changes

package/dist/index.js

@@ -7993,7 +7993,7 @@ var OpenAISkills = class {
 };
 
 // src/version.ts
-var VERSION = true ? "4.0.3" : "0.0.0-test";
+var VERSION = true ? "4.0.5" : "0.0.0-test";
 
 // src/openai-provider.ts
 function createOpenAI(options = {}) {

package/docs/03-openai.mdx

@@ -1863,13 +1863,13 @@ They support additional settings and response metadata:
 - You can use `providerOptions` to set
   - the `reasoningEffort` option (or alternatively the `reasoningEffort` model setting), which determines the amount of reasoning the model performs.
 
-- You can use response `providerMetadata` to access the number of reasoning tokens that the model generated.
+- You can use the response `usage` to access the number of reasoning tokens that the model generated via `usage.outputTokenDetails.reasoningTokens`.
 
 ```ts highlight="4,7-11,17"
 import { openai, type OpenAILanguageModelChatOptions } from '@ai-sdk/openai';
 import { generateText } from 'ai';
 
-const { text, usage, providerMetadata } = await generateText({
+const { text, usage } = await generateText({
   model: openai.chat('gpt-5'),
   prompt: 'Invent a new holiday and describe its traditions.',
   providerOptions: {
@@ -1880,10 +1880,7 @@ const { text, usage, providerMetadata } = await generateText({
 });
 
 console.log(text);
-console.log('Usage:', {
-  ...usage,
-  reasoningTokens: providerMetadata?.openai?.reasoningTokens,
-});
+console.log('Reasoning tokens:', usage.outputTokenDetails.reasoningTokens);
 ```
 
 <Note>
@@ -2215,7 +2212,7 @@ including `gpt-4o` and `gpt-4o-mini`.
 
 - Prompt caching is automatically enabled for these models, when the prompt is 1024 tokens or longer. It does
   not need to be explicitly enabled.
-- You can use response `providerMetadata` to access the number of prompt tokens that were a cache hit.
+- You can use the response `usage` to access the number of prompt tokens that were a cache hit via `usage.inputTokenDetails.cacheReadTokens`.
 - Note that caching behavior is dependent on load on OpenAI's infrastructure. Prompt prefixes generally remain in the
   cache following 5-10 minutes of inactivity before they are evicted, but during off-peak periods they may persist for up
   to an hour.
@@ -2224,15 +2221,12 @@ including `gpt-4o` and `gpt-4o-mini`.
 import { openai } from '@ai-sdk/openai';
 import { generateText } from 'ai';
 
-const { text, usage, providerMetadata } = await generateText({
+const { text, usage } = await generateText({
   model: openai.chat('gpt-4o-mini'),
   prompt: `A 1024-token or longer prompt...`,
 });
 
-console.log(`usage:`, {
-  ...usage,
-  cachedPromptTokens: providerMetadata?.openai?.cachedPromptTokens,
-});
+console.log('Cached tokens:', usage.inputTokenDetails.cacheReadTokens);
 ```
 
 To improve cache hit rates, you can manually control caching using the `promptCacheKey` option:
@@ -2241,7 +2235,7 @@ To improve cache hit rates, you can manually control caching using the `promptCa
 import { openai, type OpenAILanguageModelChatOptions } from '@ai-sdk/openai';
 import { generateText } from 'ai';
 
-const { text, usage, providerMetadata } = await generateText({
+const { text, usage } = await generateText({
   model: openai.chat('gpt-5'),
   prompt: `A 1024-token or longer prompt...`,
   providerOptions: {
@@ -2251,10 +2245,7 @@ const { text, usage, providerMetadata } = await generateText({
   },
 });
 
-console.log(`usage:`, {
-  ...usage,
-  cachedPromptTokens: providerMetadata?.openai?.cachedPromptTokens,
-});
+console.log('Cached tokens:', usage.inputTokenDetails.cacheReadTokens);
 ```
 
 For GPT-5.1 models, you can enable extended prompt caching that keeps cached prefixes active for up to 24 hours:
@@ -2263,7 +2254,7 @@ For GPT-5.1 models, you can enable extended prompt caching that keeps cached pre
 import { openai, type OpenAILanguageModelChatOptions } from '@ai-sdk/openai';
 import { generateText } from 'ai';
 
-const { text, usage, providerMetadata } = await generateText({
+const { text, usage } = await generateText({
   model: openai.chat('gpt-5.1'),
   prompt: `A 1024-token or longer prompt...`,
   providerOptions: {
@@ -2274,10 +2265,7 @@ const { text, usage, providerMetadata } = await generateText({
   },
 });
 
-console.log(`usage:`, {
-  ...usage,
-  cachedPromptTokens: providerMetadata?.openai?.cachedPromptTokens,
-});
+console.log('Cached tokens:', usage.inputTokenDetails.cacheReadTokens);
 ```
 
 #### Audio Input

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@ai-sdk/openai",
-  "version": "4.0.3",
+  "version": "4.0.5",
   "type": "module",
   "license": "Apache-2.0",
   "sideEffects": false,
@@ -35,8 +35,8 @@
     }
   },
   "dependencies": {
-    "@ai-sdk/provider": "4.0.0",
-    "@ai-sdk/provider-utils": "5.0.1"
+    "@ai-sdk/provider": "4.0.1",
+    "@ai-sdk/provider-utils": "5.0.3"
   },
   "devDependencies": {
     "@types/node": "22.19.19",