upstream-archive/stable-diffusion.cpp
1
0
Fork 0
mirror of https://github.com/leejet/stable-diffusion.cpp.git synced 2026-07-23 11:20:53 -05:00
Code Issues Packages Projects Releases Wiki Activity

feat: add hidream o1 image support (#1485)

Browse Source
This commit is contained in:
leejet
2026-05-15 00:40:21 +08:00
committed by GitHub
parent eeac950b44
commit 0665a7f8bf
20 changed files with 1703 additions and 334 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
examples/cli/README.md
View File

@@ -103,6 +103,8 @@ Generation Options:
--hires-upscaler <string> highres fix upscaler, Lanczos, Nearest, Latent, Latent (nearest), Latent
(nearest-exact), Latent (antialiased), Latent (bicubic), Latent (bicubic
antialiased), or a model name under --hires-upscalers-dir (default: Latent)
--extra-sample-args <string> extra sampler args, key=value list. Currently lcm supports noise_clip_std,
noise_scale_start, noise_scale_end
-H, --height <int> image height, in pixel space (default: 512)
-W, --width <int> image width, in pixel space (default: 512)
--steps <int> number of sample steps (default: 20)

19
examples/common/common.cpp
View File

@@ -807,6 +807,10 @@ ArgOptions SDGenerationParams::get_options() {
"Latent (antialiased), Latent (bicubic), Latent (bicubic antialiased), or a model name "
"under --hires-upscalers-dir (default: Latent)",
&hires_upscaler},
{"",
"--extra-sample-args",
"extra sampler args, key=value list. Currently lcm supports noise_clip_std, noise_scale_start, noise_scale_end",
&extra_sample_args},
};
options.int_options = {
@@ -1607,6 +1611,7 @@ bool SDGenerationParams::from_json_str(
auto parse_sample_params_json = [&](const json& sample_json,
sd_sample_params_t& target_params,
std::string& target_extra_sample_args,
std::vector<int>& target_skip_layers,
std::vector<float>* target_custom_sigmas) {
if (sample_json.contains("sample_steps") && sample_json["sample_steps"].is_number_integer()) {
@@ -1621,6 +1626,9 @@ bool SDGenerationParams::from_json_str(
if (sample_json.contains("flow_shift") && sample_json["flow_shift"].is_number()) {
target_params.flow_shift = sample_json["flow_shift"];
}
if (sample_json.contains("extra_sample_args") && sample_json["extra_sample_args"].is_string()) {
target_extra_sample_args = sample_json["extra_sample_args"].get<std::string>();
}
if (target_custom_sigmas != nullptr &&
sample_json.contains("custom_sigmas") &&
sample_json["custom_sigmas"].is_array()) {
@@ -1668,11 +1676,12 @@ bool SDGenerationParams::from_json_str(
};
if (j.contains("sample_params") && j["sample_params"].is_object()) {
parse_sample_params_json(j["sample_params"], sample_params, skip_layers, &custom_sigmas);
parse_sample_params_json(j["sample_params"], sample_params, extra_sample_args, skip_layers, &custom_sigmas);
}
if (j.contains("high_noise_sample_params") && j["high_noise_sample_params"].is_object()) {
parse_sample_params_json(j["high_noise_sample_params"],
high_noise_sample_params,
high_noise_extra_sample_args,
high_noise_skip_layers,
nullptr);
}
@@ -2099,6 +2108,8 @@ sd_img_gen_params_t SDGenerationParams::to_sd_img_gen_params_t() {
high_noise_sample_params.guidance.slg.layer_count = high_noise_skip_layers.size();
sample_params.custom_sigmas = custom_sigmas.empty() ? nullptr : custom_sigmas.data();
sample_params.custom_sigmas_count = static_cast<int>(custom_sigmas.size());
sample_params.extra_sample_args = extra_sample_args.empty() ? nullptr : extra_sample_args.c_str();
high_noise_sample_params.extra_sample_args = high_noise_extra_sample_args.empty() ? nullptr : high_noise_extra_sample_args.c_str();
cache_params.scm_mask = scm_mask.empty() ? nullptr : scm_mask.c_str();
sd_pm_params_t pm_params = {
@@ -2168,6 +2179,8 @@ sd_vid_gen_params_t SDGenerationParams::to_sd_vid_gen_params_t() {
high_noise_sample_params.guidance.slg.layer_count = high_noise_skip_layers.size();
sample_params.custom_sigmas = custom_sigmas.empty() ? nullptr : custom_sigmas.data();
sample_params.custom_sigmas_count = static_cast<int>(custom_sigmas.size());
sample_params.extra_sample_args = extra_sample_args.empty() ? nullptr : extra_sample_args.c_str();
high_noise_sample_params.extra_sample_args = high_noise_extra_sample_args.empty() ? nullptr : high_noise_extra_sample_args.c_str();
cache_params.scm_mask = scm_mask.empty() ? nullptr : scm_mask.c_str();
params.loras = lora_vec.empty() ? nullptr : lora_vec.data();
@@ -2306,6 +2319,7 @@ static json build_sampling_metadata_json(const sd_sample_params_t& sample_params
{"eta", sample_params.eta},
{"shifted_timestep", sample_params.shifted_timestep},
{"flow_shift", sample_params.flow_shift},
{"extra_sample_args", safe_json_string(sample_params.extra_sample_args)},
{"guidance",
{
{"txt_cfg", sample_params.guidance.txt_cfg},
@@ -2497,6 +2511,9 @@ std::string get_image_params(const SDContextParams& ctx_params,
}
parameter_string += "Guidance: " + std::to_string(gen_params.sample_params.guidance.distilled_guidance) + ", ";
parameter_string += "Eta: " + std::to_string(gen_params.sample_params.eta) + ", ";
if (!gen_params.extra_sample_args.empty()) {
parameter_string += "Extra sample args: " + gen_params.extra_sample_args + ", ";
}
parameter_string += "Seed: " + std::to_string(seed) + ", ";
parameter_string += "Size: " + std::to_string(gen_params.get_resolved_width()) + "x" + std::to_string(gen_params.get_resolved_height()) + ", ";
parameter_string += "Model: " + sd_basename(ctx_params.model_path) + ", ";

2
examples/common/common.h
View File

@@ -168,6 +168,8 @@ struct SDGenerationParams {
sd_sample_params_t sample_params;
sd_sample_params_t high_noise_sample_params;
std::string extra_sample_args;
std::string high_noise_extra_sample_args;
std::vector<int> skip_layers = {7, 8, 9};
std::vector<int> high_noise_skip_layers = {7, 8, 9};

2
examples/server/README.md
View File

@@ -205,6 +205,8 @@ Default Generation Options:
--hires-upscaler <string> highres fix upscaler, Lanczos, Nearest, Latent, Latent (nearest), Latent
(nearest-exact), Latent (antialiased), Latent (bicubic), Latent (bicubic
antialiased), or a model name under --hires-upscalers-dir (default: Latent)
--extra-sample-args <string> extra sampler args, key=value list. Currently lcm supports noise_clip_std,
noise_scale_start, noise_scale_end
-H, --height <int> image height, in pixel space (default: 512)
-W, --width <int> image width, in pixel space (default: 512)
--steps <int> number of sample steps (default: 20)