Skip to content

Commit 4ea75ba

Browse files
bet4itbet4it
committed
docs(grok-build): describe Responses passthrough path consistently
The Grok Build guide still referenced Chat Completions transport and its inbound translator, but grok models are now registered with api_backend = "responses" and talk to opencodex over the Responses API. Update all locales: - Replace the endpoint description with POST /v1/responses - Fix the managed-block example api_backend to "responses" - Rewrite the reasoning section to describe Responses passthrough of reasoning.summary instead of Chat Completions reasoning_content translation
1 parent 635510e commit 4ea75ba

8 files changed

Lines changed: 23 additions & 27 deletions

File tree

‎docs-site/src/content/docs/fr/guides/grok-build.md‎

Lines changed: 7 additions & 8 deletions
Original file line numberDiff line numberDiff line change
@@ -3,8 +3,8 @@ title: Grok Build
33
description: Utilisez n’importe quel modèle routé par opencodex depuis la CLI Grok Build de xAI — les modèles sont automatiquement enregistrés dans ~/.grok/config.toml pendant l’exécution du proxy.
44
---
55

6-
opencodex expose un point de terminaison compatible OpenAI `POST /v1/chat/completions` (ainsi que `/v1/responses`) sur son
7-
port local, tandis que Grok Build prend en charge les modèles personnalisés hébergés sur des serveurs compatibles OpenAI. Avec
6+
opencodex expose un point de terminaison compatible OpenAI `POST /v1/responses` sur son port local,
7+
tandis que Grok Build prend en charge les modèles personnalisés hébergés sur des serveurs compatibles OpenAI. Avec
88
cette intégration, opencodex enregistre automatiquement l’intégralité de son catalogue visible dans Grok Build :
99
aucune modification manuelle de la configuration n’est nécessaire.
1010

@@ -61,12 +61,11 @@ dans Codex. Ceux dont la liste de niveaux est vide n’affichent aucun contrôle
6161
de Codex. Les entrées GPT-5.6 natives sont distinctes : elles conservent et exposent leurs échelles de raisonnement
6262
en amont fixes, et non les métadonnées configurées pour les modèles routés.
6363

64-
Grok Build communique avec opencodex au moyen de Chat Completions et envoie `reasoning_effort` lorsque
65-
l’échelle est annoncée. Dans ce cas, le traducteur Chat Completions entrant définit par défaut le champ Responses
66-
`reasoning.summary` sur `auto` ; les traces de raisonnement parviennent donc à Grok sous la forme
67-
`delta.reasoning_content` au lieu d’être masquées. Réglez `include_reasoning: false` (ou
68-
`reasoning.summary: "none"`) si un client souhaite que le modèle réfléchisse sans renvoyer le
69-
tracé. Une valeur explicite de `reasoning.summary` prévaut lorsque les deux options sont présentes.
64+
Grok Build communique avec opencodex au moyen de l’API Responses. Lorsque la route annonce une échelle
65+
de raisonnement, la passerelle Responses transmet `reasoning.summary` telle que configurée, de sorte que
66+
les traces de raisonnement parviennent nativement à Grok sous forme d’éléments de raisonnement Responses.
67+
Réglez `reasoning.summary: "none"` si un client souhaite que le modèle réfléchisse sans renvoyer le
68+
tracé. Une valeur explicite de `reasoning.summary` prévaut sur la valeur par défaut de la route.
7069

7170
## Note d'authentification
7271

‎docs-site/src/content/docs/guides/grok-build.md‎

Lines changed: 10 additions & 11 deletions
Original file line numberDiff line numberDiff line change
@@ -3,10 +3,10 @@ title: Grok Build
33
description: Use any opencodex-routed model from xAI's Grok Build CLI — models are auto-registered into ~/.grok/config.toml while the proxy runs.
44
---
55

6-
opencodex serves an OpenAI-compatible `POST /v1/chat/completions` (and `/v1/responses`) on its
7-
local port, and Grok Build supports custom models against OpenAI-compatible servers. Starting
8-
with this integration, opencodex registers its whole visible catalog into Grok Build
9-
automatically — no manual config editing required.
6+
opencodex serves an OpenAI-compatible `POST /v1/responses` on its local port, and Grok Build
7+
supports custom models against OpenAI-compatible servers. Starting with this integration,
8+
opencodex registers its whole visible catalog into Grok Build automatically — no manual config
9+
editing required.
1010

1111
## Auto-registration
1212

@@ -18,7 +18,7 @@ into `~/.grok/config.toml`:
1818
[model.ocx-gpt-5-6-sol]
1919
model = "gpt-5.6-sol"
2020
base_url = "http://127.0.0.1:10100/v1"
21-
api_backend = "chat_completions"
21+
api_backend = "responses"
2222
api_key = "opencodex-loopback"
2323
name = "OCX gpt-5.6-sol"
2424
# ... one [model.ocx-*] table per visible model ...
@@ -61,12 +61,11 @@ in Codex. Models with an empty tier list keep no effort control, matching Codex
6161
behavior. Native GPT-5.6 entries are separate: they preserve and expose their pinned
6262
upstream reasoning ladders rather than provider-configured routed metadata.
6363

64-
Grok Build talks to opencodex over Chat Completions and sends `reasoning_effort` when
65-
the ladder is advertised. The Chat Completions inbound translator defaults the internal
66-
Responses `reasoning.summary` to `auto` in that case, so thinking traces reach Grok as
67-
`delta.reasoning_content` instead of being hidden. Set `include_reasoning: false` (or
68-
`reasoning.summary: "none"`) if a client wants the model to think without returning the
69-
trace. An explicit `reasoning.summary` wins when both knobs are present.
64+
Grok Build talks to opencodex over the Responses API. When the route advertises a reasoning
65+
ladder, the Responses passthrough forwards `reasoning.summary` as configured, so thinking
66+
traces reach Grok natively as Responses reasoning items. Set `reasoning.summary: "none"` if
67+
a client wants the model to think without returning the trace. An explicit `reasoning.summary`
68+
wins over the route default.
7069

7170
## Authentication note
7271

‎docs-site/src/content/docs/ja/guides/grok-build.md‎

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -3,7 +3,7 @@ title: グロクビルド
33
description: xAI の Grok Build CLI から opencodex でルーティングされたモデルを使用します。モデルはプロキシの実行中に ~/.grok/config.toml に自動登録されます。
44
---
55

6-
opencodex はローカル ポート上で OpenAI 互換の `POST /v1/chat/completions` (および `/v1/responses`) を提供し、Grok Build は OpenAI 互換サーバーに対するカスタム モデルをサポートします。この統合により、opencodex は表示されているカタログ全体を Grok Build に自動的に登録します。手動による構成編集は必要ありません。
6+
opencodex はローカル ポート上で OpenAI 互換の `POST /v1/responses` を提供し、Grok Build は OpenAI 互換サーバーに対するカスタム モデルをサポートします。この統合により、opencodex は表示されているカタログ全体を Grok Build に自動的に登録します。手動による構成編集は必要ありません。
77

88
## 自動登録
99

‎docs-site/src/content/docs/ko/guides/grok-build.md‎

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -3,7 +3,7 @@ title: Grok Build 안내
33
description: xAI의 Grok Build CLI에서 opencodex로 라우팅되는 모든 모델을 사용합니다. 프로세스가 실행되는 동안 모델은 `~/.grok/config.toml`에 자동 등록됩니다.
44
---
55

6-
opencodex는 로컬 포트에서 OpenAI 호환 `POST /v1/chat/completions`(및 `/v1/responses`)를 제공합니다. Grok Build는 OpenAI 호환 서버를 상대로 사용자 정의 모델을 지원합니다. 이 통합은 opencodex가 노출하는 전체 카탈로그를 Grok Build에 자동 등록합니다. 수동으로 설정 파일을 편집할 필요가 없습니다.
6+
opencodex는 로컬 포트에서 OpenAI 호환 `POST /v1/responses`를 제공합니다. Grok Build는 OpenAI 호환 서버를 상대로 사용자 정의 모델을 지원합니다. 이 통합은 opencodex가 노출하는 전체 카탈로그를 Grok Build에 자동 등록합니다. 수동으로 설정 파일을 편집할 필요가 없습니다.
77

88
## 자동 등록
99

‎docs-site/src/content/docs/ru/guides/grok-build.md‎

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -3,7 +3,7 @@ title: Grok Build
33
description: Используйте любую модель, маршрутизируемую opencodex, из CLI xAI Grok Build — пока прокси работает, модели автоматически регистрируются в ~/.grok/config.toml.
44
---
55

6-
opencodex отдаёт OpenAI-совместимый `POST /v1/chat/completions` (и `/v1/responses`) на своём
6+
opencodex отдаёт OpenAI-совместимый `POST /v1/responses` на своём
77
локальном порту, а Grok Build поддерживает custom-модели поверх OpenAI-совместимых серверов.
88
Начиная с этой интеграции, opencodex автоматически регистрирует весь свой видимый каталог в
99
Grok Build — вручную редактировать конфигурацию не нужно.

‎docs-site/src/content/docs/tr/guides/grok-build.md‎

Lines changed: 1 addition & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -3,8 +3,7 @@ title: Grok Build
33
description: xAI Grok Build CLI içerisinden opencodex ile yönlendirilen herhangi bir modeli kullanın — proxy çalışırken modeller ~/.grok/config.toml içine otomatik olarak kaydedilir.
44
---
55

6-
opencodex, yerel portunda OpenAI uyumlu bir `POST /v1/chat/completions` (ve
7-
`/v1/responses`) sunar ve Grok Build, OpenAI uyumlu sunuculara karşı özel
6+
opencodex, yerel portunda OpenAI uyumlu bir `POST /v1/responses` sunar ve Grok Build, OpenAI uyumlu sunuculara karşı özel
87
modelleri destekler. Bu entegrasyonla başlayarak opencodex, görünür kataloğunun
98
tamamını otomatik olarak Grok Build'e kaydeder — manuel yapılandırma düzenlemesi
109
gerekmez.
@@ -159,4 +158,3 @@ adlar bu nedenle noktalardan tamamen kaçınır.
159158
- **Katalog güncellemeleri:** çitle çevrili blok, enjeksiyon anındaki kataloğu
160159
yansıtır. Sağlayıcılar veya modeller ekledikten sonra yenilemek için `ocx
161160
ensure` çalıştırın (veya proxy'yi yeniden başlatın).
162-

‎docs-site/src/content/docs/zh-cn/guides/grok-build.md‎

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -3,7 +3,7 @@ title: Grok Build
33
description: 在 xAI 的 Grok Build CLI 中使用任何由 opencodex 路由的模型——在代理运行期间，模型会自动注册到 ~/.grok/config.toml。
44
---
55

6-
opencodex 在本地端口提供一个与 OpenAI 兼容的 `POST /v1/chat/completions`（以及 `/v1/responses`），而 Grok Build 支持针对与 OpenAI 兼容的服务器使用自定义模型。从这次集成开始，opencodex 会将其全部可见目录自动注册到 Grok Build 中，无需手动编辑配置。
6+
opencodex 在本地端口提供一个与 OpenAI 兼容的 `POST /v1/responses`，而 Grok Build 支持针对与 OpenAI 兼容的服务器使用自定义模型。从这次集成开始，opencodex 会将其全部可见目录自动注册到 Grok Build 中，无需手动编辑配置。
77

88
## 自动注册
99

‎docs-site/src/content/docs/zh-tw/guides/grok-build.md‎

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -3,7 +3,7 @@ title: Grok Build
33
description: 透過 xAI 的 Grok Build CLI 使用任何由 opencodex 路由的模型——代理程式執行期間會將模型自動註冊到 ~/.grok/config.toml。
44
---
55

6-
opencodex 在本機埠提供 OpenAI 相容的 `POST /v1/chat/completions`（以及 `/v1/responses`），而 Grok Build 支援對 OpenAI 相容伺服器使用自訂模型。從此整合開始，opencodex 會自動將其整個可見目錄註冊到 Grok Build——無需手動編輯設定。
6+
opencodex 在本機埠提供 OpenAI 相容的 `POST /v1/responses`，而 Grok Build 支援對 OpenAI 相容伺服器使用自訂模型。從此整合開始，opencodex 會自動將其整個可見目錄註冊到 Grok Build——無需手動編輯設定。
77

88
## 自動註冊
99

0 commit comments

Comments
 (0)