From b3ca4e088ee2fcf3ccdadd0bf5bcb6e3d9dea668 Mon Sep 17 00:00:00 2001 From: Vincent Koc Date: Thu, 23 Apr 2026 09:48:42 -0700 Subject: [PATCH] =?UTF-8?q?docs(providers):=20openai=20=E2=80=94=20drop=20?= =?UTF-8?q?stray=20anchor=20tag,=20consolidate=20trailing=20azure=20notes?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit --- docs/providers/openai.md | 25 +++++++------------------ 1 file changed, 7 insertions(+), 18 deletions(-) diff --git a/docs/providers/openai.md b/docs/providers/openai.md index c4efcfb63f0..4e9d95d2cd9 100644 --- a/docs/providers/openai.md +++ b/docs/providers/openai.md @@ -502,23 +502,14 @@ Azure portal. Azure OpenAI uses native transport and compat behavior but does not receive -OpenClaw's hidden attribution headers. See the **Native vs OpenAI-compatible -routes** accordion under [Advanced configuration](#advanced-configuration) -for details. - +OpenClaw's hidden attribution headers — see the **Native vs OpenAI-compatible +routes** accordion under [Advanced configuration](#advanced-configuration). - -For a separate Azure OpenAI Responses provider (distinct from the `openai` -provider), see the `azure-openai-responses/*` model refs in the -[Server-side compaction](#server-side-compaction-responses-api) accordion. - - - -Azure chat and Responses traffic need Azure-specific provider/API config in -addition to a base URL override. If you want Azure model calls beyond image -generation, use the onboarding flow or a provider config that sets the -appropriate Azure API/auth shape rather than assuming `openai.baseUrl` alone -is enough. +For chat or Responses traffic on Azure (beyond image generation), use the +onboarding flow or a dedicated Azure provider config — `openai.baseUrl` alone +does not pick up the Azure API/auth shape. A separate +`azure-openai-responses/*` provider exists; see +[Server-side compaction](#server-side-compaction-responses-api). ## Advanced configuration @@ -579,8 +570,6 @@ is enough. - - OpenClaw exposes a shared fast-mode toggle for both `openai/*` and `openai-codex/*`: