Skip to content

Solution for long term memory for agent coding CLIs and to facilitate handoff between different agent vendors

Rust

fix: address PR review feedback

Vitor Vilas Boase303d41cb6ec
diff --git a/CHANGELOG.md b/CHANGELOG.md
index 90136c74..5ec23cc0 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -12,7 +12,7 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
to `gemini-3.5-flash`, because Google will discontinue Gemini 2.5 Flash and
Gemini 2.5 Flash-Lite on October 20, 2026. The e2e smoke test now defaults
to `gemini-3.5-flash-lite` as its second variant. The thinking-budget
- workaround covers both the legacy models and the new defaults.
+ workaround covers both the legacy models and the new defaults. (#423)
- Documented OrcaRouter through the existing `openai-compat` provider instead
of adding a redundant provider type, including the endpoint, model, and API
key mapping needed for deployment. (#410)
diff --git a/crates/ai-memory-llm/src/gemini.rs b/crates/ai-memory-llm/src/gemini.rs
index 41fdeb63..4dabcacd 100644
--- a/crates/ai-memory-llm/src/gemini.rs
+++ b/crates/ai-memory-llm/src/gemini.rs
@@ -585,10 +585,8 @@ mod tests {
assert_eq!(prepared.get("nullable").unwrap(), &json!(true));
}
- #[test]
- fn build_request_disables_default_thinking_for_35_flash() {
- let provider =
- GeminiProvider::new(SecretString::from("test-key"), "gemini-3.5-flash").unwrap();
+ fn assert_thinking_budget_disabled(model: &str) {
+ let provider = GeminiProvider::new(SecretString::from("test-key"), model).unwrap();
let request = ChatRequest::user_prompt("emit json");
let body = serde_json::to_value(provider.build_request(&request, None)).unwrap();
assert_eq!(
@@ -597,6 +595,16 @@ mod tests {
);
}
+ #[test]
+ fn build_request_disables_default_thinking_for_25_flash() {
+ assert_thinking_budget_disabled("gemini-2.5-flash");
+ }
+
+ #[test]
+ fn build_request_disables_default_thinking_for_35_flash() {
+ assert_thinking_budget_disabled("gemini-3.5-flash");
+ }
+
#[test]
fn build_request_omits_thinking_config_for_non_flash_models() {
let provider =