From b5bd63b585b106767dc38d8daefadeecf245bcf4 Mon Sep 17 00:00:00 2001
From: dudukee <dudukee@hotmail.com>
Date: Wed, 14 Oct 2026 11:42:48 +0800
Subject: [PATCH] fix(openai): use a catalog-supported warmup reasoning effort
 (#425)

Scheduled account warmup sent reasoning.effort=none, but the Codex model
catalog lists low as the lowest supported level for every model an account
can see. Upstream answers HTTP 200 with an SSE error event (unsupported_value
on reasoning.effort), which the warmup terminal check records as
response_failed, so the cycle ends with warmed_up=0 and never starts the 5h
window.

Send low instead and cover the upstream payload with a regression test.
---
 .../openai/src/credential/quota/warmup.rs     |  6 +--
 .../openai/tests/credential/quota/warmup.rs   | 39 ++++++++++++++++++-
 2 files changed, 40 insertions(+), 5 deletions(-)

diff --git a/backend/crates/providers/openai/src/credential/quota/warmup.rs b/backend/crates/providers/openai/src/credential/quota/warmup.rs
index 9bf45da8c..2a5583df8 100644
--- a/backend/crates/providers/openai/src/credential/quota/warmup.rs
+++ b/backend/crates/providers/openai/src/credential/quota/warmup.rs
@@ -164,10 +164,8 @@ impl CodexCredentialQuotaService {
                 "service_tier".to_owned(),
                 Value::String("default".to_owned()),
             );
-            body.insert(
-                "reasoning".to_owned(),
-                serde_json::json!({"effort": "none"}),
-            );
+            // 目录里各模型的最低档都是 low，none 会收到上游 error 终态并让整次预热记为失败
+            body.insert("reasoning".to_owned(), serde_json::json!({"effort": "low"}));
             body.insert("text".to_owned(), serde_json::json!({"verbosity": "low"}));
 
             let upstream_request = CodexResponsesRequest::from_body(body);
diff --git a/backend/crates/providers/openai/tests/credential/quota/warmup.rs b/backend/crates/providers/openai/tests/credential/quota/warmup.rs
index 8aad5877b..c1c3aa3bc 100644
--- a/backend/crates/providers/openai/tests/credential/quota/warmup.rs
+++ b/backend/crates/providers/openai/tests/credential/quota/warmup.rs
@@ -1,4 +1,4 @@
-//! 验证额度预热按 SSE 终态判定结果并跳过过期凭据
+//! 验证额度预热按 SSE 终态判定结果、请求档位并跳过过期凭据
 
 use super::*;
 
@@ -40,6 +40,43 @@ async fn warmup_consumes_completed_sse_and_persists_rate_limit_event() {
     assert!(store.quota_json("acct_warmup_completed").is_some());
 }
 
+#[tokio::test]
+async fn warmup_sends_the_lowest_reasoning_effort_supported_by_the_catalog() {
+    let server = MockServer::start().await;
+    let sse = concat!(
+        "event: response.completed\n",
+        "data: {\"type\":\"response.completed\",\"response\":{\"id\":\"resp_warmup_effort\",\"status\":\"completed\",\"output\":[]}}\n\n",
+        "data: [DONE]\n\n",
+    );
+    Mock::given(method("POST"))
+        .and(path("/codex/responses"))
+        .respond_with(ResponseTemplate::new(200).set_body_raw(sse, "text/event-stream"))
+        .expect(1)
+        .mount(&server)
+        .await;
+    let store = Arc::new(MemoryAccountStore::default());
+    create_account(&store, "acct_warmup_effort").await;
+    let service = quota_service_with_base_url(
+        &store,
+        reqwest::Client::builder()
+            .no_proxy()
+            .build()
+            .expect("client"),
+        server.uri(),
+    );
+
+    let summary = service.execute_warmup("test-model").await.expect("warmup");
+
+    assert_eq!(summary.warmed_up, 1);
+    let requests = server.received_requests().await.expect("recorded requests");
+    let decoded =
+        zstd::stream::decode_all(requests[0].body.as_slice()).expect("zstd body should decode");
+    let body: serde_json::Value =
+        serde_json::from_slice(&decoded).expect("warmup body should be valid JSON");
+    // 上游按目录拒绝目录外的档位，none 曾让全部定时预热只留下 error 终态
+    assert_eq!(body["reasoning"]["effort"], "low");
+}
+
 #[tokio::test]
 async fn warmup_rejects_failed_sse_despite_successful_http_headers() {
     let server = MockServer::start().await;