Never store requests (#3212)

When item ids are sent to Responses API it will load them from the
database ignoring the provided values. This adds extra latency.

Not having the mode to store requests also allows us to simplify the
code.

## Breaking change

The `disable_response_storage` configuration option is removed.

codex-rs/core/src/client.rs

@@ -157,14 +157,6 @@ impl ModelClient {
 
         let auth_manager = self.auth_manager.clone();
 
-        let auth_mode = auth_manager
-            .as_ref()
-            .and_then(|m| m.auth())
-            .as_ref()
-            .map(|a| a.mode);
-
-        let store = prompt.store && auth_mode != Some(AuthMode::ChatGPT);
-
         let full_instructions = prompt.get_full_instructions(&self.config.model_family);
         let tools_json = create_tools_json_for_responses_api(&prompt.tools)?;
         let reasoning = create_reasoning_param_for_request(
@@ -173,9 +165,7 @@ impl ModelClient {
             self.summary,
         );
 
-        // Request encrypted COT if we are not storing responses,
-        // otherwise reasoning items will be referenced by ID
-        let include: Vec<String> = if !store && reasoning.is_some() {
+        let include: Vec<String> = if reasoning.is_some() {
             vec!["reasoning.encrypted_content".to_string()]
         } else {
             vec![]
@@ -204,7 +194,7 @@ impl ModelClient {
             tool_choice: "auto",
             parallel_tool_calls: false,
             reasoning,
-            store,
+            store: false,
             stream: true,
             include,
             prompt_cache_key: Some(self.session_id.to_string()),

codex-rs/core/src/client_common.rs

@@ -25,9 +25,6 @@ pub struct Prompt {
     /// Conversation context input items.
     pub input: Vec<ResponseItem>,
 
-    /// Whether to store response on server side (disable_response_storage = !store).
-    pub store: bool,
-
     /// Tools available to the model, including additional tools sourced from
     /// external MCP servers.
     pub(crate) tools: Vec<OpenAiTool>,
@@ -128,7 +125,6 @@ pub(crate) struct ResponsesApiRequest<'a> {
     pub(crate) tool_choice: &'static str,
     pub(crate) parallel_tool_calls: bool,
     pub(crate) reasoning: Option<Reasoning>,
-    /// true when using the Responses API.
     pub(crate) store: bool,
     pub(crate) stream: bool,
     pub(crate) include: Vec<String>,
@@ -199,7 +195,7 @@ mod tests {
             tool_choice: "auto",
             parallel_tool_calls: false,
             reasoning: None,
-            store: true,
+            store: false,
             stream: true,
             include: vec![],
             prompt_cache_key: None,
@@ -229,7 +225,7 @@ mod tests {
             tool_choice: "auto",
             parallel_tool_calls: false,
             reasoning: None,
-            store: true,
+            store: false,
             stream: true,
             include: vec![],
             prompt_cache_key: None,

codex-rs/core/src/codex.rs

@@ -184,7 +184,6 @@ impl Codex {
             base_instructions: config.base_instructions.clone(),
             approval_policy: config.approval_policy,
             sandbox_policy: config.sandbox_policy.clone(),
-            disable_response_storage: config.disable_response_storage,
             notify: config.notify.clone(),
             cwd: config.cwd.clone(),
         };
@@ -301,7 +300,6 @@ pub(crate) struct TurnContext {
     pub(crate) approval_policy: AskForApproval,
     pub(crate) sandbox_policy: SandboxPolicy,
     pub(crate) shell_environment_policy: ShellEnvironmentPolicy,
-    pub(crate) disable_response_storage: bool,
     pub(crate) tools_config: ToolsConfig,
 }
 
@@ -334,8 +332,6 @@ struct ConfigureSession {
     approval_policy: AskForApproval,
     /// How to sandbox commands executed in the system
     sandbox_policy: SandboxPolicy,
-    /// Disable server-side response storage (send full context each request)
-    disable_response_storage: bool,
 
     /// Optional external notifier command tokens. Present only when the
     /// client wants the agent to spawn a program after each completed
@@ -370,7 +366,6 @@ impl Session {
             base_instructions,
             approval_policy,
             sandbox_policy,
-            disable_response_storage,
             notify,
             cwd,
         } = configure_session;
@@ -462,7 +457,6 @@ impl Session {
             sandbox_policy,
             shell_environment_policy: config.shell_environment_policy.clone(),
             cwd,
-            disable_response_storage,
         };
         let sess = Arc::new(Session {
             session_id,
@@ -1117,7 +1111,6 @@ async fn submission_loop(
                     sandbox_policy: new_sandbox_policy.clone(),
                     shell_environment_policy: prev.shell_environment_policy.clone(),
                     cwd: new_cwd.clone(),
-                    disable_response_storage: prev.disable_response_storage,
                 };
 
                 // Install the new persistent context for subsequent tasks/turns.
@@ -1199,7 +1192,6 @@ async fn submission_loop(
                         sandbox_policy,
                         shell_environment_policy: turn_context.shell_environment_policy.clone(),
                         cwd,
-                        disable_response_storage: turn_context.disable_response_storage,
                     };
                     // TODO: record the new environment context in the conversation history
                     // no current task, spawn a new one with the per‑turn context
@@ -1604,7 +1596,6 @@ async fn run_turn(
 
     let prompt = Prompt {
         input,
-        store: !turn_context.disable_response_storage,
         tools,
         base_instructions_override: turn_context.base_instructions.clone(),
     };
@@ -1858,7 +1849,6 @@ async fn run_compact_task(
 
     let prompt = Prompt {
         input: turn_input,
-        store: !turn_context.disable_response_storage,
         tools: Vec::new(),
         base_instructions_override: Some(compact_instructions.clone()),
     };

codex-rs/core/src/config.rs

@@ -78,11 +78,6 @@ pub struct Config {
     /// Defaults to `false`.
     pub show_raw_agent_reasoning: bool,
 
-    /// Disable server-side response storage (sends the full conversation
-    /// context with every request). Currently necessary for OpenAI customers
-    /// who have opted into Zero Data Retention (ZDR).
-    pub disable_response_storage: bool,
-
     /// User-provided instructions from AGENTS.md.
     pub user_instructions: Option<String>,
 
@@ -417,11 +412,6 @@ pub struct ConfigToml {
     /// Sandbox configuration to apply if `sandbox` is `WorkspaceWrite`.
     pub sandbox_workspace_write: Option<SandboxWorkspaceWrite>,
 
-    /// Disable server-side response storage (sends the full conversation
-    /// context with every request). Currently necessary for OpenAI customers
-    /// who have opted into Zero Data Retention (ZDR).
-    pub disable_response_storage: Option<bool>,
-
     /// Optional external command to spawn for end-user notifications.
     #[serde(default)]
     pub notify: Option<Vec<String>>,
@@ -640,7 +630,6 @@ pub struct ConfigOverrides {
     pub include_plan_tool: Option<bool>,
     pub include_apply_patch_tool: Option<bool>,
     pub include_view_image_tool: Option<bool>,
-    pub disable_response_storage: Option<bool>,
     pub show_raw_agent_reasoning: Option<bool>,
     pub tools_web_search_request: Option<bool>,
 }
@@ -668,7 +657,6 @@ impl Config {
             include_plan_tool,
             include_apply_patch_tool,
             include_view_image_tool,
-            disable_response_storage,
             show_raw_agent_reasoning,
             tools_web_search_request: override_tools_web_search_request,
         } = overrides;
@@ -802,11 +790,6 @@ impl Config {
                 .unwrap_or_else(AskForApproval::default),
             sandbox_policy,
             shell_environment_policy,
-            disable_response_storage: config_profile
-                .disable_response_storage
-                .or(cfg.disable_response_storage)
-                .or(disable_response_storage)
-                .unwrap_or(false),
             notify: cfg.notify,
             user_instructions,
             base_instructions,
@@ -1071,7 +1054,6 @@ exclude_slash_tmp = true
         let toml = r#"
 model = "o3"
 approval_policy = "untrusted"
-disable_response_storage = false
 
 # Can be used to determine which profile to use if not specified by
 # `ConfigOverrides`.
@@ -1101,7 +1083,6 @@ model_provider = "openai-chat-completions"
 model = "o3"
 model_provider = "openai"
 approval_policy = "on-failure"
-disable_response_storage = true
 
 [profiles.gpt5]
 model = "gpt-5"
@@ -1199,7 +1180,6 @@ model_verbosity = "high"
                 approval_policy: AskForApproval::Never,
                 sandbox_policy: SandboxPolicy::new_read_only_policy(),
                 shell_environment_policy: ShellEnvironmentPolicy::default(),
-                disable_response_storage: false,
                 user_instructions: None,
                 notify: None,
                 cwd: fixture.cwd(),
@@ -1257,7 +1237,6 @@ model_verbosity = "high"
             approval_policy: AskForApproval::UnlessTrusted,
             sandbox_policy: SandboxPolicy::new_read_only_policy(),
             shell_environment_policy: ShellEnvironmentPolicy::default(),
-            disable_response_storage: false,
             user_instructions: None,
             notify: None,
             cwd: fixture.cwd(),
@@ -1330,7 +1309,6 @@ model_verbosity = "high"
             approval_policy: AskForApproval::OnFailure,
             sandbox_policy: SandboxPolicy::new_read_only_policy(),
             shell_environment_policy: ShellEnvironmentPolicy::default(),
-            disable_response_storage: true,
             user_instructions: None,
             notify: None,
             cwd: fixture.cwd(),
@@ -1389,7 +1367,6 @@ model_verbosity = "high"
             approval_policy: AskForApproval::OnFailure,
             sandbox_policy: SandboxPolicy::new_read_only_policy(),
             shell_environment_policy: ShellEnvironmentPolicy::default(),
-            disable_response_storage: false,
             user_instructions: None,
             notify: None,
             cwd: fixture.cwd(),

codex-rs/core/src/config_profile.rs

@@ -15,7 +15,6 @@ pub struct ConfigProfile {
     /// [`ModelProviderInfo`] to use.
     pub model_provider: Option<String>,
     pub approval_policy: Option<AskForApproval>,
-    pub disable_response_storage: Option<bool>,
     pub model_reasoning_effort: Option<ReasoningEffort>,
     pub model_reasoning_summary: Option<ReasoningSummary>,
     pub model_verbosity: Option<Verbosity>,
@@ -29,7 +28,6 @@ impl From<ConfigProfile> for codex_protocol::mcp_protocol::Profile {
             model: config_profile.model,
             model_provider: config_profile.model_provider,
             approval_policy: config_profile.approval_policy,
-            disable_response_storage: config_profile.disable_response_storage,
             model_reasoning_effort: config_profile.model_reasoning_effort,
             model_reasoning_summary: config_profile.model_reasoning_summary,
             model_verbosity: config_profile.model_verbosity,