update docs

README.md

@@ -363,6 +363,7 @@ arguments:
   --vae-tile-size [X]x[Y]            tile size for vae tiling (default: 32x32)
   --vae-relative-tile-size [X]x[Y]   relative tile size for vae tiling, in fraction of image size if < 1, in number of tiles per dim if >=1 (overrides --vae-tile-size)
   --vae-tile-overlap OVERLAP         tile overlap for vae tiling, in fraction of tile size (default: 0.5)
+  --force-sdxl-vae-conv-scale        force use of conv scale on sdxl vae
   --vae-on-cpu                       keep vae in cpu (for low vram)
   --clip-on-cpu                      keep clip in cpu (for low vram)
   --diffusion-fa                     use flash attention in the diffusion model (for low vram)

examples/cli/main.cpp

@@ -128,6 +128,7 @@ struct SDParams {
     float flow_shift         = INFINITY;
 
     sd_tiling_params_t vae_tiling_params = {false, 0, 0, 0.5f, 0.0f, 0.0f};
+    bool force_sdxl_vae_conv_scale       = false;
 
     SDParams() {
         sd_sample_params_init(&sample_params);
@@ -194,6 +195,7 @@ void print_params(SDParams params) {
     printf("    seed:                              %zd\n", params.seed);
     printf("    batch_count:                       %d\n", params.batch_count);
     printf("    vae_tiling:                        %s\n", params.vae_tiling_params.enabled ? "true" : "false");
+    printf("    force_sdxl_vae_conv_scale:         %s\n", params.force_sdxl_vae_conv_scale ? "true" : "false");
     printf("    upscale_repeats:                   %d\n", params.upscale_repeats);
     printf("    chroma_use_dit_mask:               %s\n", params.chroma_use_dit_mask ? "true" : "false");
     printf("    chroma_use_t5_mask:                %s\n", params.chroma_use_t5_mask ? "true" : "false");
@@ -287,6 +289,7 @@ void print_usage(int argc, const char* argv[]) {
     printf("  --vae-tile-size [X]x[Y]            tile size for vae tiling (default: 32x32)\n");
     printf("  --vae-relative-tile-size [X]x[Y]   relative tile size for vae tiling, in fraction of image size if < 1, in number of tiles per dim if >=1 (overrides --vae-tile-size)\n");
     printf("  --vae-tile-overlap OVERLAP         tile overlap for vae tiling, in fraction of tile size (default: 0.5)\n");
+    printf("  --force-sdxl-vae-conv-scale        force use of conv scale on sdxl vae\n");
     printf("  --vae-on-cpu                       keep vae in cpu (for low vram)\n");
     printf("  --clip-on-cpu                      keep clip in cpu (for low vram)\n");
     printf("  --diffusion-fa                     use flash attention in the diffusion model (for low vram)\n");
@@ -557,6 +560,7 @@ void parse_args(int argc, const char** argv, SDParams& params) {
 
     options.bool_options = {
         {"", "--vae-tiling", "", true, &params.vae_tiling_params.enabled},
+        {"", "--force-sdxl-vae-conv-scale", "", true, &params.force_sdxl_vae_conv_scale},
         {"", "--offload-to-cpu", "", true, &params.offload_params_to_cpu},
         {"", "--control-net-cpu", "", true, &params.control_net_cpu},
         {"", "--clip-on-cpu", "", true, &params.clip_on_cpu},
@@ -1361,6 +1365,7 @@ int main(int argc, const char* argv[]) {
         params.diffusion_flash_attn,
         params.diffusion_conv_direct,
         params.vae_conv_direct,
+        params.force_sdxl_vae_conv_scale,
         params.chroma_use_dit_mask,
         params.chroma_use_t5_mask,
         params.chroma_t5_mask_pad,

stable-diffusion.cpp

@@ -510,9 +510,13 @@ public:
                     LOG_INFO("Using Conv2d direct in the vae model");
                     first_stage_model->enable_conv2d_direct();
                 }
-                if (version == VERSION_SDXL && strlen(SAFE_STR(sd_ctx_params->vae_path)) == 0) {
+                if (version == VERSION_SDXL &&
+                    (strlen(SAFE_STR(sd_ctx_params->vae_path)) == 0 || sd_ctx_params->force_sdxl_vae_conv_scale)) {
                     float vae_conv_2d_scale = 1.f / 32.f;
-                    LOG_WARN("No VAE specified with --vae, using Conv2D scale %.3f", vae_conv_2d_scale);
+                    LOG_WARN(
+                        "No VAE specified with --vae or --force-sdxl-vae-conv-scale flag set, "
+                        "using Conv2D scale %.3f",
+                        vae_conv_2d_scale);
                     first_stage_model->set_conv2d_scale(vae_conv_2d_scale);
                 }
                 first_stage_model->alloc_params_buffer();

stable-diffusion.h

@@ -153,6 +153,7 @@ typedef struct {
     bool diffusion_flash_attn;
     bool diffusion_conv_direct;
     bool vae_conv_direct;
+    bool force_sdxl_vae_conv_scale;
     bool chroma_use_dit_mask;
     bool chroma_use_t5_mask;
     int chroma_t5_mask_pad;