diff --git a/Cargo.lock b/Cargo.lock
index 0eae257836e8d804d74e6c60bd45780a8e4f879c..e443d80906e157e5f76b6df80972e492b794c8a2 100644
--- a/Cargo.lock
+++ b/Cargo.lock
@@ -9061,7 +9061,6 @@ dependencies = [
  "editor",
  "extension",
  "extension_host",
- "feature_flags",
  "fs",
  "futures 0.3.31",
  "google_ai",
diff --git a/crates/agent_ui/src/acp/thread_view.rs b/crates/agent_ui/src/acp/thread_view.rs
index e939f053314582a42d6a009c155b243ad235802a..6431039de145a1ff762502c6230d20f0e0494271 100644
--- a/crates/agent_ui/src/acp/thread_view.rs
+++ b/crates/agent_ui/src/acp/thread_view.rs
@@ -20,10 +20,7 @@ use editor::scroll::Autoscroll;
 use editor::{
     Editor, EditorEvent, EditorMode, MultiBuffer, PathKey, SelectionEffects, SizingBehavior,
 };
-use feature_flags::{
-    AgentSharingFeatureFlag, AgentV2FeatureFlag, CloudThinkingEffortFeatureFlag,
-    FeatureFlagAppExt as _,
-};
+use feature_flags::{AgentSharingFeatureFlag, AgentV2FeatureFlag, FeatureFlagAppExt as _};
 use file_icons::FileIcons;
 use fs::Fs;
 use futures::FutureExt as _;
diff --git a/crates/agent_ui/src/acp/thread_view/active_thread.rs b/crates/agent_ui/src/acp/thread_view/active_thread.rs
index 264897b31887e5b19ec182be453c23d8b7e7802e..d1e8411fec939c6c5deef455cb08a64223b47a2a 100644
--- a/crates/agent_ui/src/acp/thread_view/active_thread.rs
+++ b/crates/agent_ui/src/acp/thread_view/active_thread.rs
@@ -2861,10 +2861,6 @@ impl AcpThreadView {
     }
 
     fn render_thinking_control(&self, cx: &mut Context<Self>) -> Option<AnyElement> {
-        if !cx.has_flag::<CloudThinkingEffortFeatureFlag>() {
-            return None;
-        }
-
         let thread = self.as_native_thread(cx)?.read(cx);
         let model = thread.model()?;
 
@@ -7205,10 +7201,6 @@ impl AcpThreadView {
     }
 
     fn cycle_thinking_effort(&mut self, cx: &mut Context<Self>) {
-        if !cx.has_flag::<CloudThinkingEffortFeatureFlag>() {
-            return;
-        }
-
         let Some(thread) = self.as_native_thread(cx) else {
             return;
         };
diff --git a/crates/feature_flags/src/flags.rs b/crates/feature_flags/src/flags.rs
index b94264879deb87b2880ef0d62ecf08489dfa8655..6fa46e48bea0b3d6b3bc482e2bfeee3eeb0e5702 100644
--- a/crates/feature_flags/src/flags.rs
+++ b/crates/feature_flags/src/flags.rs
@@ -53,15 +53,3 @@ impl FeatureFlag for DiffReviewFeatureFlag {
         false
     }
 }
-
-/// Controls whether we show the new thinking and effort level controls in the Agent Panel when using applicable models
-/// through the Zed provider (Cloud).
-pub struct CloudThinkingEffortFeatureFlag;
-
-impl FeatureFlag for CloudThinkingEffortFeatureFlag {
-    const NAME: &'static str = "cloud-thinking-effort";
-
-    fn enabled_for_staff() -> bool {
-        false
-    }
-}
diff --git a/crates/language_models/Cargo.toml b/crates/language_models/Cargo.toml
index 9685e24085495d7b028951367b1a2b4f0808c094..ece0d68152a20cbf77d0c082746959684816f115 100644
--- a/crates/language_models/Cargo.toml
+++ b/crates/language_models/Cargo.toml
@@ -34,7 +34,6 @@ credentials_provider.workspace = true
 deepseek = { workspace = true, features = ["schemars"] }
 extension.workspace = true
 extension_host.workspace = true
-feature_flags.workspace = true
 fs.workspace = true
 futures.workspace = true
 google_ai = { workspace = true, features = ["schemars"] }
diff --git a/crates/language_models/src/provider/cloud.rs b/crates/language_models/src/provider/cloud.rs
index 6caf56ebcabd1f862caf5518e33099e9160b8653..bd88d5d3b384aadef0c34e997191033045dc2de5 100644
--- a/crates/language_models/src/provider/cloud.rs
+++ b/crates/language_models/src/provider/cloud.rs
@@ -9,7 +9,6 @@ use cloud_llm_client::{
     CompletionEvent, CompletionRequestStatus, CountTokensBody, CountTokensResponse,
     ListModelsResponse, SERVER_SUPPORTS_STATUS_MESSAGES_HEADER_NAME, ZED_VERSION_HEADER_NAME,
 };
-use feature_flags::{CloudThinkingEffortFeatureFlag, FeatureFlagAppExt as _};
 use futures::{
     AsyncBufReadExt, FutureExt, Stream, StreamExt,
     future::BoxFuture,
@@ -172,26 +171,10 @@ impl State {
     }
 
     fn update_models(&mut self, response: ListModelsResponse, cx: &mut Context<Self>) {
-        let is_thinking_effort_enabled = cx.has_flag::<CloudThinkingEffortFeatureFlag>();
-
         let mut models = Vec::new();
 
         for model in response.models {
             models.push(Arc::new(model.clone()));
-
-            if !is_thinking_effort_enabled {
-                // Right now we represent thinking variants of models as separate models on the client,
-                // so we need to insert variants for any model that supports thinking.
-                if model.supports_thinking {
-                    models.push(Arc::new(cloud_llm_client::LanguageModel {
-                        id: cloud_llm_client::LanguageModelId(
-                            format!("{}-thinking", model.id).into(),
-                        ),
-                        display_name: format!("{} Thinking", model.display_name),
-                        ..model
-                    }));
-                }
-            }
         }
 
         self.default_model = models
@@ -750,13 +733,7 @@ impl LanguageModel for CloudLanguageModel {
         let intent = request.intent;
         let app_version = Some(cx.update(|cx| AppVersion::global(cx)));
         let thinking_allowed = request.thinking_allowed;
-        let is_thinking_effort_enabled =
-            cx.update(|cx| cx.has_flag::<CloudThinkingEffortFeatureFlag>());
-        let enable_thinking = if is_thinking_effort_enabled {
-            thinking_allowed && self.model.supports_thinking
-        } else {
-            thinking_allowed && self.model.id.0.ends_with("-thinking")
-        };
+        let enable_thinking = thinking_allowed && self.model.supports_thinking;
         let provider_name = provider_name(&self.model.provider);
         match self.model.provider {
             cloud_llm_client::LanguageModelProvider::Anthropic => {