From 5dd63f9426a48f413c6b40169d0df256263432fd Mon Sep 17 00:00:00 2001 From: Quintin Botes Date: Thu, 22 Oct 2026 14:17:03 +0200 Subject: [PATCH 1/2] A deployment that refuses reasoning_effort none on tool requests too is not sent none again (#1353) Signed-off-by: Quintin Botes --- crates/utopia-llm/src/lib.rs | 40 ++++++++++++++++++++++++++++++++++++ 1 file changed, 40 insertions(+) diff --git a/crates/utopia-llm/src/lib.rs b/crates/utopia-llm/src/lib.rs index aa2bfe67a..34f1c25ea 100644 --- a/crates/utopia-llm/src/lib.rs +++ b/crates/utopia-llm/src/lib.rs @@ -901,6 +901,7 @@ impl LlmClient { stream: bool, ) -> anyhow::Result { use std::sync::atomic::Ordering; + let mut retried = false; loop { let learned = self.tools_without_effort.load(Ordering::Relaxed); let resp = self @@ -923,8 +924,13 @@ impl LlmClient { { tracing::info!(model = %self.model, "端点不收带工具的推理强度，带工具的请求改发 none 并记住"); self.tools_without_effort.store(true, Ordering::Relaxed); + retried = true; continue; } + // 改发 none 也被拒：这个端点连 none 都不收，记住它只会让以后每一问都带着一个错的值 + if retried && status == reqwest::StatusCode::BAD_REQUEST { + self.tools_without_effort.store(false, Ordering::Relaxed); + } return Err(failure("LLM", status, retry_after, &parsed, &raw)); } } @@ -1694,6 +1700,40 @@ mod tests { ); } + #[tokio::test] + async fn a_deployment_that_refuses_none_too_is_not_sent_none_again() { + use wiremock::{matchers::method, Mock, MockServer, ResponseTemplate}; + let server = MockServer::start().await; + Mock::given(method("POST")) + .respond_with(ResponseTemplate::new(400).set_body_json(serde_json::json!({ "error": { + "message": "Function tools with reasoning_effort are not supported for gpt-x." } }))) + .mount(&server) + .await; + let client = + LlmClient::new(&server.uri(), None, "m").with_reasoning_effort(Some("low".into())); + let tool = serde_json::json!([{ "type": "function", "function": { "name": "ping" } }]); + let ask = [serde_json::json!({ "role": "user", "content": "x" })]; + assert!(client.chat_tools(&ask, &tool).await.is_err()); + let sent: Vec = server + .received_requests() + .await + .unwrap() + .iter() + .map(|r| serde_json::from_slice(&r.body).unwrap()) + .collect(); + assert_eq!( + sent.iter() + .map(|b| b["reasoning_effort"].clone()) + .collect::>(), + ["low", "none"] + ); + assert_eq!( + client.tools_body(&ask, Some(&tool), None, false)["reasoning_effort"], + "low", + "the flag is cleared" + ); + } + #[tokio::test] async fn another_400_on_a_tool_request_is_not_retried() { use wiremock::{matchers::method, Mock, MockServer, ResponseTemplate}; From ccb8fc37c0e94eb6d1ba0c79ef094de3a5d9a792 Mon Sep 17 00:00:00 2001 From: Quintin Botes Date: Fri, 23 Oct 2026 09:57:17 +0200 Subject: [PATCH 2/2] Only a retry refused for reasoning_effort forgets none, so an unrelated 400 keeps it. Co-Authored-By: Claude Opus 5.5 Signed-off-by: Quintin Botes --- crates/utopia-llm/src/lib.rs | 40 ++++++++++++++++++++++++++++++++++-- 1 file changed, 38 insertions(+), 2 deletions(-) diff --git a/crates/utopia-llm/src/lib.rs b/crates/utopia-llm/src/lib.rs index 34f1c25ea..9e5780c9b 100644 --- a/crates/utopia-llm/src/lib.rs +++ b/crates/utopia-llm/src/lib.rs @@ -927,8 +927,14 @@ impl LlmClient { retried = true; continue; } - // 改发 none 也被拒：这个端点连 none 都不收，记住它只会让以后每一问都带着一个错的值 - if retried && status == reqwest::StatusCode::BAD_REQUEST { + // 改发 none 也被拒，且错误还是在说推理强度：这个端点连 none 都不收，记住它只会让以后每一问都带着一个错的值。 + // 别的 400（上下文超长、工具定义不对）跟 none 无关，记住的值留着 + if retried + && status == reqwest::StatusCode::BAD_REQUEST + && err_detail(&parsed, &raw) + .to_lowercase() + .contains("reasoning_effort") + { self.tools_without_effort.store(false, Ordering::Relaxed); } return Err(failure("LLM", status, retry_after, &parsed, &raw)); @@ -1734,6 +1740,36 @@ mod tests { ); } + #[tokio::test] + async fn an_unrelated_400_on_the_retry_keeps_none() { + use wiremock::{matchers::method, Mock, MockServer, ResponseTemplate}; + let server = MockServer::start().await; + Mock::given(method("POST")) + .respond_with(ResponseTemplate::new(400).set_body_json(serde_json::json!({ "error": { + "message": "Function tools with reasoning_effort are not supported for gpt-x." } }))) + .up_to_n_times(1) + .mount(&server) + .await; + Mock::given(method("POST")) + .respond_with( + ResponseTemplate::new(400).set_body_json(serde_json::json!({ "error": { + "message": "This model's maximum context length is 8192 tokens." } })), + ) + .mount(&server) + .await; + let client = + LlmClient::new(&server.uri(), None, "m").with_reasoning_effort(Some("low".into())); + let tool = serde_json::json!([{ "type": "function", "function": { "name": "ping" } }]); + let ask = [serde_json::json!({ "role": "user", "content": "x" })]; + assert!(client.chat_tools(&ask, &tool).await.is_err()); + assert_eq!(server.received_requests().await.unwrap().len(), 2); + assert_eq!( + client.tools_body(&ask, Some(&tool), None, false)["reasoning_effort"], + "none", + "the flag is kept" + ); + } + #[tokio::test] async fn another_400_on_a_tool_request_is_not_retried() { use wiremock::{matchers::method, Mock, MockServer, ResponseTemplate};