Model: merge split models when converting

Author: stduhpf

examples/cli/main.cpp

@@ -952,7 +952,7 @@ int main(int argc, const char* argv[]) {
     }
 
     if (params.mode == CONVERT) {
-        bool success = convert(params.model_path.c_str(), params.vae_path.c_str(), params.output_path.c_str(), params.wtype);
+        bool success = convert(params.model_path.c_str(), params.clip_l_path.c_str(), params.clip_g_path.c_str(), params.t5xxl_path.c_str(), params.diffusion_model_path.c_str(), params.vae_path.c_str(), params.output_path.c_str(), params.wtype);
         if (!success) {
             fprintf(stderr,
                     "convert '%s'/'%s' to '%s' failed\n",

model.cpp

@@ -2076,12 +2076,41 @@ void setConvertImatrixCollector(void* collector) {
     imatrix_collector = ((IMatrixCollector*)collector);
 }
 
-bool convert(const char* input_path, const char* vae_path, const char* output_path, sd_type_t output_type) {
+bool convert(const char* model_path, const char* clip_l_path, const char* clip_g_path, const char* t5xxl_path, const char* diffusion_model_path, const char* vae_path, const char* output_path, sd_type_t output_type) {
     ModelLoader model_loader;
 
-    if (!model_loader.init_from_file(input_path)) {
-        LOG_ERROR("init model loader from file failed: '%s'", input_path);
-        return false;
+    if (model_path != NULL && strlen(model_path) > 0) {
+        if (!model_loader.init_from_file(model_path)) {
+            LOG_ERROR("init model loader from file failed: '%s'", model_path);
+            return false;
+        }
+    }
+
+    if (clip_l_path != NULL && strlen(clip_l_path) > 0) {
+        if (!model_loader.init_from_file(clip_l_path, "text_encoders.clip_l.transformer.")) {
+            LOG_ERROR("init model loader from file failed: '%s'", clip_l_path);
+            return false;
+        }
+    }
+
+    if (clip_g_path != NULL && strlen(clip_g_path) > 0) {
+        if (!model_loader.init_from_file(clip_g_path, "text_encoders.clip_g.transformer.")) {
+            LOG_ERROR("init model loader from file failed: '%s'", clip_g_path);
+            return false;
+        }
+    }
+    if (t5xxl_path != NULL && strlen(t5xxl_path) > 0) {
+        if (!model_loader.init_from_file(t5xxl_path, "text_encoders.t5xxl.transformer.")) {
+            LOG_ERROR("init model loader from file failed: '%s'", t5xxl_path);
+            return false;
+        }
+    }
+
+    if (diffusion_model_path != NULL && strlen(diffusion_model_path) > 0) {
+        if (!model_loader.init_from_file(diffusion_model_path, "model.diffusion_model.")) {
+            LOG_ERROR("init model loader from file failed: '%s'", diffusion_model_path);
+            return false;
+        }
     }
 
     if (vae_path != NULL && strlen(vae_path) > 0) {

stable-diffusion.h

@@ -253,7 +253,7 @@ SD_API void free_upscaler_ctx(upscaler_ctx_t* upscaler_ctx);
 SD_API sd_image_t upscale(upscaler_ctx_t* upscaler_ctx, sd_image_t input_image, uint32_t upscale_factor);
 
 SD_API void setConvertImatrixCollector(void * collector);
-SD_API bool convert(const char* input_path, const char* vae_path, const char* output_path, enum sd_type_t output_type);
+SD_API bool convert(const char* model_path, const char* clip_l_path, const char* clip_g_path, const char* t5xxl_path, const char* diffusion_model_path, const char* vae_path, const char* output_path, enum sd_type_t output_type);
 
 SD_API uint8_t* preprocess_canny(uint8_t* img,
                                  int width,