From bd65eeb0450502dbcd2d26e29d673d3730c53be5 Mon Sep 17 00:00:00 2001
From: Diego Rodrigues de Sa e Souza
 <8016841+diegosouzapw@users.noreply.github.com>
Date: Mon, 6 Jul 2026 19:24:08 -0300
Subject: [PATCH] fix(executors): strip client_metadata for NVIDIA requests
 (port from 9router#1887) (#6411)

strip client_metadata for NVIDIA (port #1887) (net +1/-0, test OK). Integrated into release/v3.8.46.
---
 CHANGELOG.md                                      |  1 +
 open-sse/executors/default.ts                     |  9 ++++++---
 ...executor-default-strip-client-metadata.test.ts | 15 +++++++++++++++
 3 files changed, 22 insertions(+), 3 deletions(-)

diff --git a/CHANGELOG.md b/CHANGELOG.md
index 5211b391e0..01d60ea0b0 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -21,6 +21,7 @@
 
 ### 🐛 Bug Fixes
 
+- **fix(executors):** strip the OpenAI-Codex/Claude-CLI `client_metadata` passthrough field for **NVIDIA** requests. NVIDIA's OpenAI-compatible wrapper rejects it with `400 Unsupported parameter`, the same class already handled for `cerebras`/`mistral`; `nvidia` (executor `default`) was missing from the strip allowlist so Codex/Claude-Code passthrough requests 400'd. Regression guard: `tests/unit/executor-default-strip-client-metadata.test.ts` (+nvidia case). (thanks @phidinhmanh)
 - **fix(translator):** strip the Claude-style `thinking` field for **NVIDIA `z-ai/glm-5.2`**. NVIDIA's OpenAI-compatible wrapper 400s on `thinking` (a Claude-format client routed here leaves a `thinking:{type:"adaptive"}`); the existing strip rule only dropped `reasoning`. Same class already handled for `minimax-m2.7`. Regression guard: `tests/unit/nvidia-minimax-thinking-strip.test.ts` (+glm-5.2 case). (thanks @phidinhmanh)
 - **fix(translator):** suppress the streamed `</think>` close marker for the **Antigravity IDE** client. On thinking-only turns Antigravity rendered a bare `</think>` as the sole visible content, tripping its loop-detection and wasting requests. Antigravity's UA (`vscode/<v> (Antigravity/<v>)`) is added to the marker-suppress allowlist (alongside OpenCode); Claude Code / Cursor still get the marker, and `x-omniroute-thinking-marker: on` force-restores it. Regression guard: `tests/unit/think-close-marker-suppress-5245.test.ts`. (thanks @abdofallah)
 - **fix(executors):** strip nested `reasoning_content` from messages for **Mistral**. Mistral's API returns `422 extra_forbidden` when an assistant message carries `reasoning_content` (replayed thinking from a prior turn, e.g. via the Codex `/responses` path); the generic top-level 400 field-downgrade retry never covered the nested per-message field. `DefaultExecutor` now strips it for provider `mistral` only, so DeepSeek (which requires replayed `reasoning_content`) is unaffected. Regression guard: `tests/unit/mistral-strip-reasoning-content-1649.test.ts`. (thanks @xxy9468615)
diff --git a/open-sse/executors/default.ts b/open-sse/executors/default.ts
index c643a9e65e..b2697a349e 100644
--- a/open-sse/executors/default.ts
+++ b/open-sse/executors/default.ts
@@ -549,8 +549,9 @@ export class DefaultExecutor extends BaseExecutor {
     withDefaults = this.defaultResponsesTextFormat(withDefaults);
 
     // Port of decolua/9router commit d652300e:
-    // Cerebras returns 400 (wrong_api_format) and Mistral returns 422
-    // (extra_forbidden) when the forwarded body carries `client_metadata`
+    // Cerebras returns 400 (wrong_api_format), Mistral returns 422
+    // (extra_forbidden), and NVIDIA's OpenAI-compatible wrapper returns 400
+    // (Unsupported parameter) when the forwarded body carries `client_metadata`
     // (an OpenAI Codex / Claude CLI passthrough field with no equivalent on
     // these upstreams). Strip it before sending downstream. Other providers
     // (notably `openai` / `codex`) intentionally keep it.
@@ -558,7 +559,9 @@ export class DefaultExecutor extends BaseExecutor {
       withDefaults &&
       typeof withDefaults === "object" &&
       !Array.isArray(withDefaults) &&
-      (this.provider === "cerebras" || this.provider === "mistral") &&
+      (this.provider === "cerebras" ||
+        this.provider === "mistral" ||
+        this.provider === "nvidia") &&
       Object.prototype.hasOwnProperty.call(withDefaults, "client_metadata")
     ) {
       const withoutClientMetadata = { ...(withDefaults as Record<string, unknown>) };
diff --git a/tests/unit/executor-default-strip-client-metadata.test.ts b/tests/unit/executor-default-strip-client-metadata.test.ts
index 5430991806..b8b48ab4dc 100644
--- a/tests/unit/executor-default-strip-client-metadata.test.ts
+++ b/tests/unit/executor-default-strip-client-metadata.test.ts
@@ -54,6 +54,21 @@ test("DefaultExecutor.transformRequest strips client_metadata for mistral", () =
   );
 });
 
+test("DefaultExecutor.transformRequest strips client_metadata for nvidia (port from 9router#1887)", () => {
+  const executor = new DefaultExecutor("nvidia");
+  const out = executor.transformRequest(
+    "any",
+    bodyWithClientMetadata(),
+    STREAM,
+    CREDENTIALS
+  ) as Record<string, unknown>;
+  assert.equal(
+    Object.prototype.hasOwnProperty.call(out, "client_metadata"),
+    false,
+    "nvidia forward body must not contain client_metadata"
+  );
+});
+
 test("DefaultExecutor.transformRequest preserves client_metadata for other providers", () => {
   // Sanity check: the strip must be scoped, not global. Openai keeps it
   // (codex flow needs it), the cerebras/mistral strip is the only carve-out.