Public/koboldcpp
1
0
Fork 0
mirror of https://github.com/LostRuins/koboldcpp.git synced 2026-09-19 01:05:09 +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/model.cpp
+7 -17
View File
@@ -46,19 +46,6 @@
#define ST_HEADER_SIZE_LEN 8
static std::string format(const char* fmt, ...) {
va_list ap;
va_list ap2;
va_start(ap, fmt);
va_copy(ap2, ap);
int size = vsnprintf(NULL, 0, fmt, ap);
std::vector<char> buf(size + 1);
int size2 = vsnprintf(buf.data(), size + 1, fmt, ap2);
va_end(ap2);
va_end(ap);
return std::string(buf.data(), size);
}
uint64_t read_u64(uint8_t* buffer) {
// little endian
uint64_t value = 0;
@@ -295,8 +282,8 @@ void convert_tensor(void* src,
} else {
auto qtype = ggml_get_type_traits(src_type);
if (qtype->to_float == nullptr) {
throw std::runtime_error(format("type %s unsupported for integer quantization: no dequantization available",
ggml_type_name(src_type)));
throw std::runtime_error(sd_format("type %s unsupported for integer quantization: no dequantization available",
ggml_type_name(src_type)));
}
qtype->to_float(src, (float*)dst, n);
}
@@ -305,8 +292,8 @@ void convert_tensor(void* src,
// src_type is quantized => dst_type == GGML_TYPE_F16 or dst_type is quantized
auto qtype = ggml_get_type_traits(src_type);
if (qtype->to_float == nullptr) {
throw std::runtime_error(format("type %s unsupported for integer quantization: no dequantization available",
ggml_type_name(src_type)));
throw std::runtime_error(sd_format("type %s unsupported for integer quantization: no dequantization available",
ggml_type_name(src_type)));
}
std::vector<char> buf;
buf.resize(sizeof(float) * n);
@@ -1125,6 +1112,9 @@ SDVersion ModelLoader::get_sd_version() {
if (tensor_storage.name.find("model.diffusion_model.double_stream_modulation_img.lin.weight") != std::string::npos) {
return VERSION_FLUX2;
}
if (tensor_storage.name.find("model.diffusion_model.cap_embedder.0.weight") != std::string::npos) {
return VERSION_Z_IMAGE;
}
if (tensor_storage.name.find("model.diffusion_model.blocks.0.cross_attn.norm_k.weight") != std::string::npos) {
is_wan = true;
}