add qwen image t2i pipeline

examples/cli/main.cpp

@@ -27,8 +27,6 @@
 
 #include "avi_writer.h"
 
-#include "qwen_image.hpp"
-
 #if defined(_WIN32)
 #define NOMINMAX
 #include <windows.h>
@@ -61,6 +59,7 @@ struct SDParams {
     std::string clip_g_path;
     std::string clip_vision_path;
     std::string t5xxl_path;
+    std::string qwen2vl_path;
     std::string diffusion_model_path;
     std::string high_noise_diffusion_model_path;
     std::string vae_path;
@@ -146,6 +145,7 @@ void print_params(SDParams params) {
     printf("    clip_g_path:                       %s\n", params.clip_g_path.c_str());
     printf("    clip_vision_path:                  %s\n", params.clip_vision_path.c_str());
     printf("    t5xxl_path:                        %s\n", params.t5xxl_path.c_str());
+    printf("    qwen2vl_path:                      %s\n", params.qwen2vl_path.c_str());
     printf("    diffusion_model_path:              %s\n", params.diffusion_model_path.c_str());
     printf("    high_noise_diffusion_model_path:   %s\n", params.high_noise_diffusion_model_path.c_str());
     printf("    vae_path:                          %s\n", params.vae_path.c_str());
@@ -217,6 +217,7 @@ void print_usage(int argc, const char* argv[]) {
     printf("  --clip_g                           path to the clip-g text encoder\n");
     printf("  --clip_vision                      path to the clip-vision encoder\n");
     printf("  --t5xxl                            path to the t5xxl text encoder\n");
+    printf("  --qwen2vl                          path to the qwen2vl text encoder\n");
     printf("  --vae [VAE]                        path to vae\n");
     printf("  --taesd [TAESD_PATH]               path to taesd. Using Tiny AutoEncoder for fast decoding (low quality)\n");
     printf("  --control-net [CONTROL_PATH]       path to control net model\n");
@@ -486,6 +487,7 @@ void parse_args(int argc, const char** argv, SDParams& params) {
         {"", "--clip_g", "", &params.clip_g_path},
         {"", "--clip_vision", "", &params.clip_vision_path},
         {"", "--t5xxl", "", &params.t5xxl_path},
+        {"", "--qwen2vl", "", &params.qwen2vl_path},
         {"", "--diffusion-model", "", &params.diffusion_model_path},
         {"", "--high-noise-diffusion-model", "", &params.high_noise_diffusion_model_path},
         {"", "--vae", "", &params.vae_path},
@@ -945,7 +947,7 @@ std::string get_image_params(SDParams params, int64_t seed) {
         parameter_string += " " + std::string(sd_schedule_name(params.sample_params.scheduler));
     }
     parameter_string += ", ";
-    for (const auto& te : {params.clip_l_path, params.clip_g_path, params.t5xxl_path}) {
+    for (const auto& te : {params.clip_l_path, params.clip_g_path, params.t5xxl_path, params.qwen2vl_path}) {
         if (!te.empty()) {
             parameter_string += "TE: " + sd_basename(te) + ", ";
         }
@@ -1140,10 +1142,6 @@ bool load_images_from_dir(const std::string dir,
 
 int main(int argc, const char* argv[]) {
     SDParams params;
-    params.verbose = true;
-    sd_set_log_callback(sd_log_cb, (void*)&params);
-    Qwen::QwenImageRunner::load_from_file_and_test(argv[1]);
-    exit(1);
     parse_args(argc, argv, params);
     params.sample_params.guidance.slg.layers                 = params.skip_layers.data();
     params.sample_params.guidance.slg.layer_count            = params.skip_layers.size();
@@ -1323,6 +1321,7 @@ int main(int argc, const char* argv[]) {
         params.clip_g_path.c_str(),
         params.clip_vision_path.c_str(),
         params.t5xxl_path.c_str(),
+        params.qwen2vl_path.c_str(),
         params.diffusion_model_path.c_str(),
         params.high_noise_diffusion_model_path.c_str(),
         params.vae_path.c_str(),