Public/koboldcpp
1
0
Fork 0
mirror of https://github.com/LostRuins/koboldcpp.git synced 2026-09-18 08:44:57 +02:00
Code Issues Packages Projects Releases Wiki Activity

sd: sync to master-387-e4c50f1

Browse Source
This commit is contained in:
Wagner Bruna
2025-12-01 18:19:41 -03:00
parent e971eaefe3
commit fcb86f22df
12 changed files with 992 additions and 49 deletions
Download Patch File Download Diff File Expand all files Collapse all files
otherarch/sdcpp/stable-diffusion.cpp
+32 -10
View File
@@ -47,6 +47,7 @@ const char* model_version_to_str[] = {
"Wan 2.2 TI2V",
"Qwen Image",
"Flux.2",
"Z-Image",
};
const char* sampling_methods_str[] = {
@@ -498,7 +499,7 @@ public:
} else if (sd_version_is_sd3(version)) {
scale_factor = 1.5305f;
shift_factor = 0.0609f;
} else if (sd_version_is_flux(version)) {
} else if (sd_version_is_flux(version) || sd_version_is_z_image(version)) {
scale_factor = 0.3611f;
shift_factor = 0.1159f;
} else if (sd_version_is_wan(version) ||
@@ -616,6 +617,16 @@ public:
tensor_storage_map,
"model.diffusion_model",
version);
} else if (sd_version_is_z_image(version)) {
cond_stage_model = std::make_shared<LLMEmbedder>(clip_backend,
offload_params_to_cpu,
tensor_storage_map,
version);
diffusion_model = std::make_shared<ZImageModel>(backend,
offload_params_to_cpu,
tensor_storage_map,
"model.diffusion_model",
version);
} else { // SD1.x SD2.x SDXL
if (strstr(SAFE_STR(sd_ctx_params->photo_maker_path), "v2")) {
cond_stage_model = std::make_shared<FrozenCLIPEmbedderWithCustomWords>(clip_backend,
@@ -993,6 +1004,13 @@ public:
shift = 3.0;
}
denoiser = std::make_shared<DiscreteFlowDenoiser>(shift);
} else if (sd_version_is_z_image(version)) {
LOG_INFO("running in FLOW mode");
float shift = sd_ctx_params->flow_shift;
if (shift == INFINITY) {
shift = 3.0f;
}
denoiser = std::make_shared<DiscreteFlowDenoiser>(shift);
} else if (is_using_v_parameterization) {
LOG_INFO("running in v-prediction mode");
denoiser = std::make_shared<CompVisVDenoiser>();
@@ -1508,7 +1526,7 @@ public:
if (sd_version_is_sd3(version)) {
latent_rgb_proj = sd3_latent_rgb_proj;
latent_rgb_bias = sd3_latent_rgb_bias;
} else if (sd_version_is_flux(version)) {
} else if (sd_version_is_flux(version) || sd_version_is_z_image(version)) {
latent_rgb_proj = flux_latent_rgb_proj;
latent_rgb_bias = flux_latent_rgb_bias;
} else if (sd_version_is_wan(version) || sd_version_is_qwen_image(version)) {
@@ -1809,6 +1827,8 @@ public:
shifted_t = std::max((int64_t)0, std::min((int64_t)(TIMESTEPS - 1), shifted_t));
LOG_DEBUG("shifting timestep from %.2f to %" PRId64 " (sigma: %.4f)", t, shifted_t, sigma);
timesteps_vec.assign(1, (float)shifted_t);
} else if (sd_version_is_z_image(version)) {
timesteps_vec.assign(1, 1000.f - t);
} else {
timesteps_vec.assign(1, t);
}
@@ -2247,8 +2267,9 @@ public:
ggml_tensor* vae_encode(ggml_context* work_ctx, ggml_tensor* x, bool encode_video = false) {
int64_t t0 = ggml_time_ms();
ggml_tensor* result = nullptr;
int W = x->ne[0] / get_vae_scale_factor();
int H = x->ne[1] / get_vae_scale_factor();
const int vae_scale_factor = get_vae_scale_factor();
int W = x->ne[0] / vae_scale_factor;
int H = x->ne[1] / vae_scale_factor;
int C = get_latent_channel();
if (vae_tiling_params.enabled && !encode_video) {
// TODO wan2.2 vae support?
@@ -2284,7 +2305,7 @@ public:
auto on_tiling = [&](ggml_tensor* in, ggml_tensor* out, bool init) {
first_stage_model->compute(n_threads, in, false, &out, work_ctx);
};
sd_tiling_non_square(x, result, 8, tile_size_x, tile_size_y, tile_overlap, on_tiling);
sd_tiling_non_square(x, result, vae_scale_factor, tile_size_x, tile_size_y, tile_overlap, on_tiling);
} else {
first_stage_model->compute(n_threads, x, false, &result, work_ctx);
}
@@ -2295,7 +2316,7 @@ public:
auto on_tiling = [&](ggml_tensor* in, ggml_tensor* out, bool init) {
tae_first_stage->compute(n_threads, in, false, &out, nullptr);
};
sd_tiling(x, result, 8, 64, 0.5f, on_tiling);
sd_tiling(x, result, vae_scale_factor, 64, 0.5f, on_tiling);
} else {
tae_first_stage->compute(n_threads, x, false, &result, work_ctx);
}
@@ -2371,8 +2392,9 @@ public:
}
ggml_tensor* decode_first_stage(ggml_context* work_ctx, ggml_tensor* x, bool decode_video = false) {
int64_t W = x->ne[0] * get_vae_scale_factor();
int64_t H = x->ne[1] * get_vae_scale_factor();
const int vae_scale_factor = get_vae_scale_factor();
int64_t W = x->ne[0] * vae_scale_factor;
int64_t H = x->ne[1] * vae_scale_factor;
int64_t C = 3;
ggml_tensor* result = nullptr;
if (decode_video) {
@@ -2412,7 +2434,7 @@ public:
auto on_tiling = [&](ggml_tensor* in, ggml_tensor* out, bool init) {
first_stage_model->compute(n_threads, in, true, &out, nullptr);
};
sd_tiling_non_square(x, result, 8, tile_size_x, tile_size_y, tile_overlap, on_tiling);
sd_tiling_non_square(x, result, vae_scale_factor, tile_size_x, tile_size_y, tile_overlap, on_tiling);
} else {
first_stage_model->compute(n_threads, x, true, &result, work_ctx);
}
@@ -2424,7 +2446,7 @@ public:
auto on_tiling = [&](ggml_tensor* in, ggml_tensor* out, bool init) {
tae_first_stage->compute(n_threads, in, true, &out);
};
sd_tiling(x, result, 8, 64, 0.5f, on_tiling);
sd_tiling(x, result, vae_scale_factor, 64, 0.5f, on_tiling);
} else {
tae_first_stage->compute(n_threads, x, true, &result);
}