docs: include_usage streaming contract changed: no separate usage-only chunk; usage rides on the final content chunk

content/.metadata.json

@@ -107,11 +107,11 @@
     },
     "content/en/news/news1226.md": {
       "url": "https://api-docs.deepseek.com/news/news1226",
-      "error": "served fallback shell for https://api-docs.deepseek.com/news/news1226 (canonical: https://api-docs.deepseek.com/)"
+      "title": "🚀 Introducing DeepSeek-V3"
     },
     "content/en/news/news250325.md": {
       "url": "https://api-docs.deepseek.com/news/news250325",
-      "error": "served fallback shell for https://api-docs.deepseek.com/news/news250325 (canonical: https://api-docs.deepseek.com/)"
+      "title": "DeepSeek-V3-0324 Release"
     },
     "content/en/news/news0725.md": {
       "url": "https://api-docs.deepseek.com/news/news0725",

content/en/api/create-chat-completion.md

@@ -2,7 +2,7 @@
 title: "Chat Completions API"
 description: "Creates a model response for the given chat conversation."
 source: https://api-docs.deepseek.com/api/create-chat-completion
-fetched: 2026-08-23
+fetched: 2026-08-26
 ---
 
 # Chat Completions API
@@ -260,7 +260,9 @@ Options for streaming response. Only set this when you set `stream: true`.
 
 **include\_usage** boolean
 
-If set, an additional chunk will be streamed before the `data: [DONE]` message. The `usage` field on this chunk shows the token usage statistics for the entire request, and the `choices` field will always be an empty array. All other chunks will also include a `usage` field, but with a null value.
+If set to `true`, all chunks in the stream will include a `usage` field, whose value is `null` on every chunk except the last one. If omitted or set to `false`, the `usage` field is absent from all chunks except the last one.
+
+Either way, the last chunk before the `data: [DONE]` message carries the token usage statistics for the entire request in its `usage` field. Note that no separate usage-only chunk is emitted: the statistics ride on the last content chunk, whose `choices` array always contains exactly one element that carries no new content and a non-null `finish_reason`.
 
 **temperature** numbernullable
 

content/en/api/create-completion.md

@@ -2,7 +2,7 @@
 title: "FIM Completion API (Beta)"
 description: "FIM (Fill In the Middle) Completion API.<br/>User must set `base_url='https://api.deepseek.com/beta'` to use this feature."
 source: https://api-docs.deepseek.com/api/create-completion
-fetched: 2026-08-02
+fetched: 2026-08-26
 ---
 
 # FIM Completion API (Beta)
@@ -79,7 +79,9 @@ Options for streaming response. Only set this when you set `stream: true`.
 
 **include\_usage** boolean
 
-If set, an additional chunk will be streamed before the `data: [DONE]` message. The `usage` field on this chunk shows the token usage statistics for the entire request, and the `choices` field will always be an empty array. All other chunks will also include a `usage` field, but with a null value.
+If set to `true`, all chunks in the stream will include a `usage` field, whose value is `null` on every chunk except the last one. If omitted or set to `false`, the `usage` field is absent from all chunks except the last one.
+
+Either way, the last chunk before the `data: [DONE]` message carries the token usage statistics for the entire request in its `usage` field. Note that no separate usage-only chunk is emitted: the statistics ride on the last content chunk, whose `choices` array always contains exactly one element that carries no new content and a non-null `finish_reason`.
 
 **suffix** stringnullable
 

content/zh-cn/api/create-chat-completion.md

@@ -2,7 +2,7 @@
 title: "Chat Completions API"
 description: "根据输入的上下文，来让模型补全对话内容。"
 source: https://api-docs.deepseek.com/zh-cn/api/create-chat-completion
-fetched: 2026-08-23
+fetched: 2026-08-26
 ---
 
 # Chat Completions API
@@ -258,7 +258,9 @@ string
 
 **include\_usage** boolean
 
-如果设置为 true，在流式消息最后的 `data: [DONE]` 之前将会传输一个额外的块。此块上的 usage 字段显示整个请求的 token 使用统计信息，而 choices 字段将始终是一个空数组。所有其他块也将包含一个 usage 字段，但其值为 null。
+如果设置为 `true`，流式返回的所有块都会包含 `usage` 字段，其中除最后一个块外，该字段的值均为 `null`。如果不设置或设置为 `false`，则除最后一个块外，其余块都不含 `usage` 字段。
+
+无论是否设置，`data: [DONE]` 之前的最后一个块都会在其 `usage` 字段中给出整个请求的 token 使用统计信息。请注意，这里不会单独下发一个只含 usage 的块：统计信息附加在最后一个内容块上，该块的 `choices` 数组始终只包含一个元素，其中不含新增内容且 `finish_reason` 非 null。
 
 **temperature** numbernullable
 

content/zh-cn/api/create-completion.md

@@ -2,7 +2,7 @@
 title: "FIM 补全 API（Beta）"
 description: "FIM (Fill In the Middle) 补全 API。<br/>用户需要设置 `base_url='https://api.deepseek.com/beta'` 来使用此功能。"
 source: https://api-docs.deepseek.com/zh-cn/api/create-completion
-fetched: 2026-08-23
+fetched: 2026-08-26
 ---
 
 # FIM 补全 API（Beta）
@@ -79,7 +79,9 @@ string
 
 **include\_usage** boolean
 
-如果设置为 true，在流式消息最后的 `data: [DONE]` 之前将会传输一个额外的块。此块上的 usage 字段显示整个请求的 token 使用统计信息，而 choices 字段将始终是一个空数组。所有其他块也将包含一个 usage 字段，但其值为 null。
+如果设置为 `true`，流式返回的所有块都会包含 `usage` 字段，其中除最后一个块外，该字段的值均为 `null`。如果不设置或设置为 `false`，则除最后一个块外，其余块都不含 `usage` 字段。
+
+无论是否设置，`data: [DONE]` 之前的最后一个块都会在其 `usage` 字段中给出整个请求的 token 使用统计信息。请注意，这里不会单独下发一个只含 usage 的块：统计信息附加在最后一个内容块上，该块的 `choices` 数组始终只包含一个元素，其中不含新增内容且 `finish_reason` 非 null。
 
 **suffix** stringnullable
 

llms-full.txt

@@ -7,7 +7,7 @@
 title: "Chat Completions API"
 description: "Creates a model response for the given chat conversation."
 source: https://api-docs.deepseek.com/api/create-chat-completion
-fetched: 2026-08-23
+fetched: 2026-08-26
 ---
 
 # Chat Completions API
@@ -265,7 +265,9 @@ Options for streaming response. Only set this when you set `stream: true`.
 
 **include\_usage** boolean
 
-If set, an additional chunk will be streamed before the `data: [DONE]` message. The `usage` field on this chunk shows the token usage statistics for the entire request, and the `choices` field will always be an empty array. All other chunks will also include a `usage` field, but with a null value.
+If set to `true`, all chunks in the stream will include a `usage` field, whose value is `null` on every chunk except the last one. If omitted or set to `false`, the `usage` field is absent from all chunks except the last one.
+
+Either way, the last chunk before the `data: [DONE]` message carries the token usage statistics for the entire request in its `usage` field. Note that no separate usage-only chunk is emitted: the statistics ride on the last content chunk, whose `choices` array always contains exactly one element that carries no new content and a non-null `finish_reason`.
 
 **temperature** numbernullable
 
@@ -990,7 +992,7 @@ Loading...
 title: "FIM Completion API (Beta)"
 description: "FIM (Fill In the Middle) Completion API.<br/>User must set `base_url='https://api.deepseek.com/beta'` to use this feature."
 source: https://api-docs.deepseek.com/api/create-completion
-fetched: 2026-08-02
+fetched: 2026-08-26
 ---
 
 # FIM Completion API (Beta)
@@ -1067,7 +1069,9 @@ Options for streaming response. Only set this when you set `stream: true`.
 
 **include\_usage** boolean
 
-If set, an additional chunk will be streamed before the `data: [DONE]` message. The `usage` field on this chunk shows the token usage statistics for the entire request, and the `choices` field will always be an empty array. All other chunks will also include a `usage` field, but with a null value.
+If set to `true`, all chunks in the stream will include a `usage` field, whose value is `null` on every chunk except the last one. If omitted or set to `false`, the `usage` field is absent from all chunks except the last one.
+
+Either way, the last chunk before the `data: [DONE]` message carries the token usage statistics for the entire request in its `usage` field. Note that no separate usage-only chunk is emitted: the statistics ride on the last content chunk, whose `choices` array always contains exactly one element that carries no new content and a non-null `finish_reason`.
 
 **suffix** stringnullable