openai · copyberry · Jul 10, 2026 · Jul 10, 2026
diff --git a/codex-rs/app-server/tests/common/models_cache.rs b/codex-rs/app-server/tests/common/models_cache.rs
@@ -35,7 +35,6 @@ fn preset_to_info(preset: &ModelPreset, priority: i32) -> ModelInfo {
         base_instructions: "base instructions".to_string(),
         model_messages: None,
         include_skills_usage_instructions: false,
-        supports_reasoning_summaries: false,
         default_reasoning_summary: ReasoningSummary::Auto,
         support_verbosity: false,
         default_verbosity: None,

diff --git a/codex-rs/app-server/tests/suite/v2/model_list.rs b/codex-rs/app-server/tests/suite/v2/model_list.rs
@@ -186,7 +186,6 @@ async fn list_models_uses_chatgpt_remote_catalog_as_source_of_truth() -> Result<
         "priority": 0,
         "upgrade": null,
         "base_instructions": "base instructions",
-        "supports_reasoning_summaries": false,
         "support_verbosity": false,
         "default_verbosity": null,
         "apply_patch_tool_type": null,

diff --git a/codex-rs/codex-api/src/endpoint/models.rs b/codex-rs/codex-api/src/endpoint/models.rs
@@ -209,7 +209,6 @@ mod tests {
                     "priority": 1,
                     "upgrade": null,
                     "base_instructions": "base instructions",
-                    "supports_reasoning_summaries": false,
                     "support_verbosity": false,
                     "default_verbosity": null,
                     "apply_patch_tool_type": null,

diff --git a/codex-rs/codex-api/tests/models_integration.rs b/codex-rs/codex-api/tests/models_integration.rs
@@ -81,7 +81,6 @@ async fn models_client_hits_models_endpoint() {
             base_instructions: "base instructions".to_string(),
             model_messages: None,
             include_skills_usage_instructions: false,
-            supports_reasoning_summaries: false,
             default_reasoning_summary: ReasoningSummary::Auto,
             support_verbosity: false,
             default_verbosity: None,

diff --git a/codex-rs/config/src/config_toml.rs b/codex-rs/config/src/config_toml.rs
@@ -355,9 +355,6 @@ pub struct ConfigToml {
     /// Optional verbosity control for GPT-5 models (Responses API `text.verbosity`).
     pub model_verbosity: Option<Verbosity>,
 
-    /// Override to force-enable reasoning summaries for the configured model.
-    pub model_supports_reasoning_summaries: Option<bool>,
-
     /// Optional path to a JSON model catalog (applied on startup only).
     /// Per-thread `config` overrides are accepted but do not reapply this (no-ops).
     pub model_catalog_json: Option<AbsolutePathBuf>,

diff --git a/codex-rs/core/config.schema.json b/codex-rs/core/config.schema.json
@@ -5292,10 +5292,6 @@
     "model_reasoning_summary": {
       "$ref": "#/definitions/ReasoningSummary"
     },
-    "model_supports_reasoning_summaries": {
-      "description": "Override to force-enable reasoning summaries for the configured model.",
-      "type": "boolean"
-    },
     "model_verbosity": {
       "allOf": [
         {

diff --git a/codex-rs/core/src/client.rs b/codex-rs/core/src/client.rs
@@ -804,25 +804,21 @@ impl ModelClient {
         model_info: &ModelInfo,
         effort: Option<ReasoningEffortConfig>,
         summary: ReasoningSummaryConfig,
-    ) -> Option<Reasoning> {
-        if model_info.supports_reasoning_summaries {
-            Some(Reasoning {
-                effort: effort
-                    .or_else(|| model_info.default_reasoning_level.clone())
-                    .map(reasoning_effort_for_request),
-                summary: if summary == ReasoningSummaryConfig::None {
-                    None
-                } else {
-                    Some(summary)
-                },
-                // When Responses Lite is disabled, omit context so Responses uses the default,
-                // which is currently `current_turn`.
-                context: model_info
-                    .use_responses_lite
-                    .then_some(ReasoningContext::AllTurns),
-            })
-        } else {
-            None
+    ) -> Reasoning {
+        Reasoning {
+            effort: effort
+                .or_else(|| model_info.default_reasoning_level.clone())
+                .map(reasoning_effort_for_request),
+            summary: if summary == ReasoningSummaryConfig::None {
+                None
+            } else {
+                Some(summary)
+            },
+            // When Responses Lite is disabled, omit context so Responses uses the default,
+            // which is currently `current_turn`.
+            context: model_info
+                .use_responses_lite
+                .then_some(ReasoningContext::AllTurns),
         }
     }
 
@@ -871,12 +867,8 @@ impl ModelClient {
             .then_some(StreamOptions {
                 reasoning_summary_delivery: codex_api::ReasoningSummaryDelivery::SequentialCutoff,
             });
-        let reasoning = Self::build_reasoning(model_info, effort, summary);
-        let include = if reasoning.is_some() {
-            vec!["reasoning.encrypted_content".to_string()]
-        } else {
-            Vec::new()
-        };
+        let reasoning = Some(Self::build_reasoning(model_info, effort, summary));
+        let include = vec!["reasoning.encrypted_content".to_string()];
         let verbosity = if model_info.support_verbosity {
             self.state.model_verbosity.or(model_info.default_verbosity)
         } else {

diff --git a/codex-rs/core/src/client_tests.rs b/codex-rs/core/src/client_tests.rs
@@ -262,7 +262,6 @@ fn test_model_info() -> ModelInfo {
         "upgrade": null,
         "base_instructions": "base instructions",
         "model_messages": null,
-        "supports_reasoning_summaries": false,
         "support_verbosity": false,
         "default_verbosity": null,
         "apply_patch_tool_type": null,

diff --git a/codex-rs/core/src/config/mod.rs b/codex-rs/core/src/config/mod.rs
@@ -952,9 +952,6 @@ pub struct Config {
     /// using the Responses API. When unset, the model catalog default is used.
     pub model_reasoning_summary: Option<ReasoningSummary>,
 
-    /// Optional override to force-enable reasoning summaries for the configured model.
-    pub model_supports_reasoning_summaries: Option<bool>,
-
     /// Optional full model catalog loaded from `model_catalog_json`.
     /// When set, this replaces the bundled catalog for the current process.
     pub model_catalog: Option<ModelsResponse>,
@@ -1476,7 +1473,6 @@ impl Config {
             tool_output_token_limit: self.tool_output_token_limit,
             base_instructions: self.base_instructions.clone(),
             personality_enabled: self.features.enabled(Feature::Personality),
-            model_supports_reasoning_summaries: self.model_supports_reasoning_summaries,
             model_catalog: self.model_catalog.clone(),
         }
     }
@@ -3903,7 +3899,6 @@ impl Config {
             model_reasoning_effort: cfg.model_reasoning_effort,
             plan_mode_reasoning_effort: cfg.plan_mode_reasoning_effort,
             model_reasoning_summary: cfg.model_reasoning_summary,
-            model_supports_reasoning_summaries: cfg.model_supports_reasoning_summaries,
             model_catalog,
             model_verbosity: cfg.model_verbosity,
             chatgpt_base_url: cfg

diff --git a/codex-rs/core/src/guardian/review_session.rs b/codex-rs/core/src/guardian/review_session.rs
@@ -720,14 +720,10 @@ async fn run_review_on_session(
             &params.spawn_config.to_models_manager_config(),
         )
         .await;
-    let guardian_reasoning_effort = if model_info.supports_reasoning_summaries {
-        params
-            .reasoning_effort
-            .clone()
-            .or_else(|| model_info.default_reasoning_level.clone())
-    } else {
-        None
-    };
+    let guardian_reasoning_effort = params
+        .reasoning_effort
+        .clone()
+        .or_else(|| model_info.default_reasoning_level.clone());
     let mut analytics_result =
         GuardianReviewAnalyticsResult::from_session(GuardianReviewSessionAnalyticsParams {
             guardian_thread_id: review_session.codex.session.thread_id.to_string(),

diff --git a/codex-rs/core/src/session/turn_context.rs b/codex-rs/core/src/session/turn_context.rs
@@ -177,13 +177,9 @@ impl TurnContext {
     }
 
     pub(crate) fn effective_reasoning_effort(&self) -> Option<ReasoningEffortConfig> {
-        if self.model_info.supports_reasoning_summaries {
-            self.reasoning_effort
-                .clone()
-                .or_else(|| self.model_info.default_reasoning_level.clone())
-        } else {
-            None
-        }
+        self.reasoning_effort
+            .clone()
+            .or_else(|| self.model_info.default_reasoning_level.clone())
     }
 
     pub(crate) fn effective_reasoning_effort_for_tracing(&self) -> String {

diff --git a/codex-rs/core/tests/responses_headers.rs b/codex-rs/core/tests/responses_headers.rs
@@ -348,7 +348,6 @@ async fn responses_respects_model_info_overrides_from_config() {
     config.model = Some("gpt-3.5-turbo".to_string());
     config.model_provider_id = provider.name.clone();
     config.model_provider = provider.clone();
-    config.model_supports_reasoning_summaries = Some(true);
     config.model_reasoning_summary = Some(ReasoningSummary::Detailed);
     let effort = config.model_reasoning_effort.clone();
     let summary = config.model_reasoning_summary;

diff --git a/codex-rs/core/tests/suite/auto_review.rs b/codex-rs/core/tests/suite/auto_review.rs
@@ -245,7 +245,6 @@ fn remote_model_with_auto_review_override(slug: &str, review_model: &str) -> Mod
         base_instructions: "base instructions".to_string(),
         model_messages: None,
         include_skills_usage_instructions: false,
-        supports_reasoning_summaries: false,
         default_reasoning_summary: ReasoningSummary::Auto,
         support_verbosity: false,
         default_verbosity: None,

diff --git a/codex-rs/core/tests/suite/client.rs b/codex-rs/core/tests/suite/client.rs
@@ -2539,7 +2539,6 @@ async fn user_turn_explicit_reasoning_summary_overrides_model_catalog_default()
         .iter_mut()
         .find(|model| model.slug == "gpt-5.4")
         .expect("gpt-5.4 exists in bundled models.json");
-    model.supports_reasoning_summaries = true;
     model.default_reasoning_summary = ReasoningSummary::Detailed;
 
     let TestCodex {
@@ -2662,7 +2661,6 @@ async fn reasoning_summary_none_overrides_model_catalog_default() -> anyhow::Res
         .iter_mut()
         .find(|model| model.slug == "gpt-5.4")
         .expect("gpt-5.4 exists in bundled models.json");
-    model.supports_reasoning_summaries = true;
     model.default_reasoning_summary = ReasoningSummary::Detailed;
 
     let TestCodex { codex, .. } = test_codex()

diff --git a/codex-rs/core/tests/suite/client_websockets.rs b/codex-rs/core/tests/suite/client_websockets.rs
@@ -681,8 +681,7 @@ async fn responses_websocket_sends_responses_lite_metadata_per_request() {
     .await;
 
     let harness = websocket_harness(&server).await;
-    let mut normal_model_info = harness.model_info.clone();
-    normal_model_info.supports_reasoning_summaries = true;
+    let normal_model_info = harness.model_info.clone();
     let mut lite_model_info = normal_model_info.clone();
     lite_model_info.use_responses_lite = true;
     let mut session = harness.client.new_session();

diff --git a/codex-rs/core/tests/suite/model_switching.rs b/codex-rs/core/tests/suite/model_switching.rs
@@ -98,7 +98,6 @@ fn test_model_info(
         base_instructions: "base instructions".to_string(),
         model_messages: None,
         include_skills_usage_instructions: false,
-        supports_reasoning_summaries: false,
         default_reasoning_summary: ReasoningSummary::Auto,
         support_verbosity: false,
         default_verbosity: None,
@@ -926,7 +925,6 @@ async fn model_switch_to_smaller_model_updates_token_context_window() -> Result<
         base_instructions: "base instructions".to_string(),
         model_messages: None,
         include_skills_usage_instructions: false,
-        supports_reasoning_summaries: false,
         default_reasoning_summary: ReasoningSummary::Auto,
         support_verbosity: false,
         default_verbosity: None,

diff --git a/codex-rs/core/tests/suite/models_cache_ttl.rs b/codex-rs/core/tests/suite/models_cache_ttl.rs
@@ -368,7 +368,6 @@ fn test_remote_model(slug: &str, priority: i32) -> ModelInfo {
         base_instructions: "base instructions".to_string(),
         model_messages: None,
         include_skills_usage_instructions: false,
-        supports_reasoning_summaries: false,
         default_reasoning_summary: ReasoningSummary::Auto,
         support_verbosity: false,
         default_verbosity: None,

diff --git a/codex-rs/core/tests/suite/multi_agent_mode.rs b/codex-rs/core/tests/suite/multi_agent_mode.rs
@@ -27,7 +27,6 @@ const PROACTIVE_TEXT: &str = "Proactive multi-agent delegation is active.";
 const CUSTOM_MODE_HINT_TEXT: &str = "Use the configured delegation policy.";
 
 fn add_ultra_reasoning(model_info: &mut ModelInfo) {
-    model_info.supports_reasoning_summaries = true;
     model_info
         .supported_reasoning_levels
         .push(ReasoningEffortPreset {

diff --git a/codex-rs/core/tests/suite/personality.rs b/codex-rs/core/tests/suite/personality.rs
@@ -566,7 +566,6 @@ async fn remote_model_friendly_personality_instructions_with_feature() -> anyhow
             approvals: None,
         }),
         include_skills_usage_instructions: false,
-        supports_reasoning_summaries: false,
         default_reasoning_summary: ReasoningSummary::Auto,
         support_verbosity: false,
         default_verbosity: None,
@@ -683,7 +682,6 @@ async fn user_turn_personality_remote_model_template_includes_update_message() -
             approvals: None,
         }),
         include_skills_usage_instructions: false,
-        supports_reasoning_summaries: false,
         default_reasoning_summary: ReasoningSummary::Auto,
         support_verbosity: false,
         default_verbosity: None,

diff --git a/codex-rs/core/tests/suite/remote_models.rs b/codex-rs/core/tests/suite/remote_models.rs
@@ -346,7 +346,6 @@ async fn remote_models_long_model_slug_is_sent_with_custom_reasoning() -> Result
             description: custom_reasoning_effort.to_string(),
         },
     ];
-    remote_model.supports_reasoning_summaries = true;
     remote_model.default_reasoning_summary = ReasoningSummary::Detailed;
     mount_models_once(
         &server,
@@ -491,7 +490,6 @@ async fn remote_models_remote_model_uses_unified_exec() -> Result<()> {
         base_instructions: "base instructions".to_string(),
         model_messages: None,
         include_skills_usage_instructions: false,
-        supports_reasoning_summaries: false,
         default_reasoning_summary: ReasoningSummary::Auto,
         support_verbosity: false,
         default_verbosity: None,
@@ -745,7 +743,6 @@ async fn remote_models_apply_remote_base_instructions() -> Result<()> {
         base_instructions: remote_base.to_string(),
         model_messages: None,
         include_skills_usage_instructions: false,
-        supports_reasoning_summaries: false,
         default_reasoning_summary: ReasoningSummary::Auto,
         support_verbosity: false,
         default_verbosity: None,
@@ -1272,7 +1269,6 @@ fn test_remote_model_with_policy(
         base_instructions: "base instructions".to_string(),
         model_messages: None,
         include_skills_usage_instructions: false,
-        supports_reasoning_summaries: false,
         default_reasoning_summary: ReasoningSummary::Auto,
         support_verbosity: false,
         default_verbosity: None,

diff --git a/codex-rs/core/tests/suite/rmcp_client.rs b/codex-rs/core/tests/suite/rmcp_client.rs
@@ -1653,7 +1653,6 @@ async fn stdio_image_responses_are_sanitized_for_text_only_model() -> anyhow::Re
                 base_instructions: "base instructions".to_string(),
                 model_messages: None,
                 include_skills_usage_instructions: false,
-                supports_reasoning_summaries: false,
                 default_reasoning_summary: ReasoningSummary::Auto,
                 support_verbosity: false,
                 default_verbosity: None,

diff --git a/codex-rs/core/tests/suite/spawn_agent_description.rs b/codex-rs/core/tests/suite/spawn_agent_description.rs
@@ -72,7 +72,6 @@ fn test_model_info(
         base_instructions: "base instructions".to_string(),
         model_messages: None,
         include_skills_usage_instructions: false,
-        supports_reasoning_summaries: false,
         default_reasoning_summary: ReasoningSummary::Auto,
         support_verbosity: false,
         default_verbosity: None,

diff --git a/codex-rs/core/tests/suite/view_image.rs b/codex-rs/core/tests/suite/view_image.rs
@@ -1375,7 +1375,6 @@ async fn view_image_tool_returns_unsupported_message_for_text_only_model() -> an
         base_instructions: "base instructions".to_string(),
         model_messages: None,
         include_skills_usage_instructions: false,
-        supports_reasoning_summaries: false,
         default_reasoning_summary: ReasoningSummary::Auto,
         support_verbosity: false,
         default_verbosity: None,

diff --git a/codex-rs/model-provider/src/provider.rs b/codex-rs/model-provider/src/provider.rs
@@ -421,7 +421,6 @@ mod tests {
             "priority": 0,
             "upgrade": null,
             "base_instructions": "base instructions",
-            "supports_reasoning_summaries": false,
             "support_verbosity": false,
             "default_verbosity": null,
             "apply_patch_tool_type": null,

diff --git a/codex-rs/models-manager/models.json b/codex-rs/models-manager/models.json
diff --git a/codex-rs/models-manager/src/config.rs b/codex-rs/models-manager/src/config.rs
@@ -7,6 +7,5 @@ pub struct ModelsManagerConfig {
     pub tool_output_token_limit: Option<usize>,
     pub base_instructions: Option<String>,
     pub personality_enabled: bool,
-    pub model_supports_reasoning_summaries: Option<bool>,
     pub model_catalog: Option<ModelsResponse>,
 }
diff --git a/codex-rs/models-manager/src/manager_tests.rs b/codex-rs/models-manager/src/manager_tests.rs
@@ -51,7 +51,6 @@ fn remote_model_with_visibility(
             "priority": priority,
             "upgrade": null,
             "base_instructions": "base instructions",
-            "supports_reasoning_summaries": false,
             "support_verbosity": false,
             "default_verbosity": null,
             "apply_patch_tool_type": null,

diff --git a/codex-rs/models-manager/src/model_info.rs b/codex-rs/models-manager/src/model_info.rs
@@ -21,11 +21,6 @@ const LOCAL_PRAGMATIC_TEMPLATE: &str = "You are a deeply pragmatic, effective so
 const PERSONALITY_PLACEHOLDER: &str = "{{ personality }}";
 
 pub fn with_config_overrides(mut model: ModelInfo, config: &ModelsManagerConfig) -> ModelInfo {
-    if let Some(supports_reasoning_summaries) = config.model_supports_reasoning_summaries
-        && supports_reasoning_summaries
-    {
-        model.supports_reasoning_summaries = true;
-    }
     if let Some(context_window) = config.model_context_window {
         model.context_window = Some(
             model
@@ -93,7 +88,6 @@ pub fn model_info_from_slug(slug: &str) -> ModelInfo {
         base_instructions: BASE_INSTRUCTIONS.to_string(),
         model_messages: local_personality_messages_for_slug(slug),
         include_skills_usage_instructions: false,
-        supports_reasoning_summaries: false,
         default_reasoning_summary: ReasoningSummary::Auto,
         support_verbosity: false,
         default_verbosity: None,