diff --git a/gpt4all-backend/llamamodel.cpp b/gpt4all-backend/llamamodel.cpp
index 4a61f782..7fd9fce1 100644
--- a/gpt4all-backend/llamamodel.cpp
+++ b/gpt4all-backend/llamamodel.cpp
@@ -303,11 +303,11 @@ bool LLamaModel::initializeGPUDevice(const LLModel::GPUDevice &device, std::stri
     vkDevice.vendor = device.vendor;
     result = ggml_vk_init_device(vkDevice);
     if (!result && unavail_reason) {
-        *unavail_reason = "failed to init device";
+        *unavail_reason = "failed to init GPU";
     }
 #else
     if (unavail_reason) {
-        *unavail_reason = "built without kompute";
+        *unavail_reason = "built without Kompute";
     }
 #endif
     return result;
diff --git a/gpt4all-backend/llmodel.h b/gpt4all-backend/llmodel.h
index ad33c82d..3001281b 100644
--- a/gpt4all-backend/llmodel.h
+++ b/gpt4all-backend/llmodel.h
@@ -99,7 +99,7 @@ public:
     virtual bool initializeGPUDevice(size_t /*memoryRequired*/, const std::string& /*device*/) { return false; }
     virtual bool initializeGPUDevice(const GPUDevice &/*device*/, std::string *unavail_reason = nullptr) {
         if (unavail_reason) {
-            *unavail_reason = "unsupported model type";
+            *unavail_reason = "model has no GPU support";
         }
         return false;
     }
diff --git a/gpt4all-chat/chatllm.cpp b/gpt4all-chat/chatllm.cpp
index 1b950829..9325b763 100644
--- a/gpt4all-chat/chatllm.cpp
+++ b/gpt4all-chat/chatllm.cpp
@@ -287,7 +287,7 @@ bool ChatLLM::loadModel(const ModelInfo &modelInfo)
                     if (!device) {
                         // GPU not available
                     } else if (!m_llModelInfo.model->initializeGPUDevice(*device, &unavail_reason)) {
-                        emit reportFallbackReason(QString::fromStdString("<br>Using CPU: " + unavail_reason));
+                        emit reportFallbackReason(QString::fromStdString("<br>" + unavail_reason));
                     } else {
                         actualDevice = QString::fromStdString(device->name);
                     }
@@ -302,14 +302,14 @@ bool ChatLLM::loadModel(const ModelInfo &modelInfo)
                 } else if (!success) {
                     // llama_init_from_file returned nullptr
                     emit reportDevice("CPU");
-                    emit reportFallbackReason("<br>Using CPU: loading failed (out of VRAM?)");
+                    emit reportFallbackReason("<br>GPU loading failed (out of VRAM?)");
                     success = m_llModelInfo.model->loadModel(filePath.toStdString());
                 } else if (!m_llModelInfo.model->usingGPUDevice()) {
                     // ggml_vk_init was not called in llama.cpp
                     // We might have had to fallback to CPU after load if the model is not possible to accelerate
                     // for instance if the quantization method is not supported on Vulkan yet
                     emit reportDevice("CPU");
-                    emit reportFallbackReason("<br>Using CPU: unsupported model or quant");
+                    emit reportFallbackReason("<br>model or quant has no GPU support");
                 }
 
                 MySettings::globalInstance()->setAttemptModelLoad(QString());