diff --git a/HANDOFF_DILIGENT.md b/HANDOFF_DILIGENT.md
index 46df8ff2..f614d927 100644
--- a/HANDOFF_DILIGENT.md
+++ b/HANDOFF_DILIGENT.md
@@ -161,6 +161,7 @@ Verified locally on Turnip Adreno 750:
   - `GenerateMipmap` (Diligent GPU mip generation on state textures)
   - `GetTexImage` / `GetTextureImage` (RGBA8 readback)
   - Fence sync entries (`FenceSync` / `ClientWaitSync` / `WaitSync` / `DeleteSync` / `GetSyncStatus`) as CPU always-signaled fallback
+  - Timer query entries (`BeginTimeElapsedQuery` / `EndTimeElapsedQuery` / `QueryCounterTimestamp` / `GetQueryResult64` etc.) as CPU `steady_clock` fallback
   - `ReadPixels` from default and user color attachments
 - Primitive expansion:
   - `GL_TRIANGLE_FAN` expanded to triangle list
@@ -219,7 +220,7 @@ Notes:
 - Textures auto-sync `ITextureObject` → Diligent resources, including mip levels and sampler state; compressed textures and integer/3-channel formats that Diligent lacks are still skipped.
 - Global UBO (default-block `glUniform*`) and named application UBO blocks (through `glBindBufferBase`/`glUniformBlockBinding`) now upload and bind; SSBOs are still not fed from frontend buffer bindings.
 - No swapchain / EGL window surface presentation yet; `Present()` only flushes.
-- No transform feedback / GPU queries / non-color readback; fence sync is provided as a CPU always-signaled fallback.
+- No transform feedback / GPU-accelerated queries / non-color readback; fence sync and timer queries use CPU fallbacks.
 - Draw range, multi-draw, instanced-draw wrappers, clear-buffer, blit, read-pixels, CopyTexImage*, CopyImageSubData, GenerateMipmap, GetTexImage/GetTextureImage and indirect draws are now wired; buffer subdata paths still remain.
 - A last-PSO cache now avoids recreating the pipeline when program/render-state/topology/VAO layout is unchanged; texture/UBO resources are still rebound dynamically per draw.
 - The `GLFunctionsTable` is only partially populated.
diff --git a/MobileGL/MG_Backend/Diligent/BackendObject_Diligent.cpp b/MobileGL/MG_Backend/Diligent/BackendObject_Diligent.cpp
index 3536a38e..00f64ed1 100644
--- a/MobileGL/MG_Backend/Diligent/BackendObject_Diligent.cpp
+++ b/MobileGL/MG_Backend/Diligent/BackendObject_Diligent.cpp
@@ -17,6 +17,7 @@
 #include <DeviceContext.h>
 
 #include <exception>
+#include <chrono>
 
 namespace MobileGL::MG_Backend::DiligentBackend {
     namespace {
@@ -50,6 +51,12 @@ namespace MobileGL::MG_Backend::DiligentBackend {
             Uint32 BaseInstance = 0;
         };
 
+        struct CpuTimerQuery {
+            std::chrono::steady_clock::time_point Start;
+            Uint64 TimestampNs = 0;
+            Bool Available = false;
+        };
+
         const Uint8* ResolveIndirectCommandBytes(const void* indirect, SizeT requiredBytes, const char* label) {
             auto drawBuffer = MG_State::pGLContext->GetBufferBindingSlot(BufferTarget::DrawIndirect).GetBoundObject();
             if (drawBuffer) {
@@ -571,6 +578,57 @@ namespace MobileGL::MG_Backend::DiligentBackend {
             }
         }
 
+        Bool IsTimerQuerySupported() {
+            return true;
+        }
+
+        BackendQueryHandle BeginTimeElapsedQuery() {
+            auto* query = new CpuTimerQuery;
+            query->Start = std::chrono::steady_clock::now();
+            query->Available = false;
+            return query;
+        }
+
+        void EndTimeElapsedQuery(BackendQueryHandle query) {
+            if (query == nullptr) {
+                return;
+            }
+            auto* cpuQuery = static_cast<CpuTimerQuery*>(query);
+            const auto now = std::chrono::steady_clock::now();
+            cpuQuery->TimestampNs = static_cast<Uint64>(
+                std::chrono::duration_cast<std::chrono::nanoseconds>(now - cpuQuery->Start).count());
+            cpuQuery->Available = true;
+        }
+
+        BackendQueryHandle QueryCounterTimestamp() {
+            auto* query = new CpuTimerQuery;
+            query->TimestampNs = static_cast<Uint64>(
+                std::chrono::duration_cast<std::chrono::nanoseconds>(
+                    std::chrono::steady_clock::now().time_since_epoch()).count());
+            query->Available = true;
+            return query;
+        }
+
+        Bool IsQueryResultAvailable(BackendQueryHandle query) {
+            return query != nullptr && static_cast<CpuTimerQuery*>(query)->Available;
+        }
+
+        Bool GetQueryResult64(BackendQueryHandle query, Bool wait, Uint64* outNanoseconds) {
+            if (query == nullptr || outNanoseconds == nullptr) {
+                return false;
+            }
+            auto* cpuQuery = static_cast<CpuTimerQuery*>(query);
+            if (!cpuQuery->Available && !wait) {
+                return false;
+            }
+            *outNanoseconds = cpuQuery->TimestampNs;
+            return true;
+        }
+
+        void DeleteBackendQuery(BackendQueryHandle query) {
+            delete static_cast<CpuTimerQuery*>(query);
+        }
+
         BackendSyncHandle FenceSync() {
             // CPU fallback fence: always signaled is a valid implementation for a
             // backend without native sync primitives. The handle still round-trips
@@ -724,6 +782,13 @@ namespace MobileGL::MG_Backend::DiligentBackend {
         m_functions.GL.WaitSync = WaitSync;
         m_functions.GL.DeleteSync = DeleteSync;
         m_functions.GL.GetSyncStatus = GetSyncStatus;
+        m_functions.GL.IsTimerQuerySupported = IsTimerQuerySupported;
+        m_functions.GL.BeginTimeElapsedQuery = BeginTimeElapsedQuery;
+        m_functions.GL.EndTimeElapsedQuery = EndTimeElapsedQuery;
+        m_functions.GL.QueryCounterTimestamp = QueryCounterTimestamp;
+        m_functions.GL.IsQueryResultAvailable = IsQueryResultAvailable;
+        m_functions.GL.GetQueryResult64 = GetQueryResult64;
+        m_functions.GL.DeleteBackendQuery = DeleteBackendQuery;
         m_functions.Present = Present;
 
         m_initialized = true;