Public/koboldcpp
1
0
Fork 0
mirror of https://github.com/LostRuins/koboldcpp.git synced 2026-09-19 17:25:07 +02:00
Code Issues Packages Projects Releases Wiki Activity

Merge commit 'd646c9d15500a702425e8a90c19d2400deecbd0c' into concedo_experimental

Browse Source 
# Conflicts:
#	.github/actions/windows-setup-rocm/action.yml
#	.github/workflows/build-apple.yml
#	.github/workflows/release.yml
#	.github/workflows/server-self-hosted.yml
#	examples/training/README.md
#	ggml/src/ggml-hexagon/ggml-hexagon.cpp
#	ggml/src/ggml-hexagon/htp/hvx-arith.h
#	ggml/src/ggml-hexagon/htp/hvx-log.h
#	ggml/src/ggml-hexagon/htp/hvx-norm.h
#	ggml/src/ggml-hexagon/htp/hvx-scale.h
#	ggml/src/ggml-hexagon/htp/hvx-sqrt.h
#	ggml/src/ggml-hexagon/htp/unary-ops.c
#	ggml/src/ggml-hexagon/htp/unary-ops.h
#	ggml/src/ggml-sycl/mmvq.cpp
#	ggml/src/ggml-sycl/vecdotq.hpp
#	tests/test-backend-ops.cpp
#	tests/test-mtmd-c-api.c
#	tests/test-mtmd-impl.cpp
This commit is contained in:
Concedo
2026-09-04 16:05:40 +08:00
parent 99be3d0990 d646c9d155
commit a3e6006530
78 changed files with 693 additions and 221 deletions
Download Patch File Download Diff File Expand all files Collapse all files
tools/server/server-common.cpp
+12 -11
View File
@@ -1062,8 +1062,7 @@ json oaicompat_completion_params_parse(const json & body) {
static void handle_media(
std::vector<raw_buffer> & out_files,
const std::string & url,
const std::string & media_path,
bool accept_base64_uri) {
const std::string & media_path) {
if (!media_path.empty()) {
// should already be enforced by arg.cpp, but checking just in case
GGML_ASSERT(media_path.back() == DIRECTORY_SEPARATOR);
@@ -1104,15 +1103,17 @@ static void handle_media(
data.assign((std::istreambuf_iterator<char>(file)), std::istreambuf_iterator<char>());
out_files.push_back(data);
} else if (accept_base64_uri && string_starts_with(url, "data:")) {
// try to decode base64 image
} else if (string_starts_with(url, "data:")) {
// try to decode base64 image, video, or audio
std::vector<std::string> parts = string_split<std::string>(url, /*separator*/ ',');
if (parts.size() != 2) {
throw std::runtime_error("Invalid uri-encoded base64 value");
} else if (!string_starts_with(parts[0], "data:image/")) {
throw std::runtime_error("Invalid uri format: " + parts[0]);
throw std::invalid_argument("Invalid uri-encoded base64 value");
} else if (!string_starts_with(parts[0], "data:image/")
&& !string_starts_with(parts[0], "data:video/")
&& !string_starts_with(parts[0], "data:audio/")) {
throw std::invalid_argument("Invalid uri format: " + parts[0]);
} else if (!string_ends_with(parts[0], "base64")) {
throw std::runtime_error("uri must be base64 encoded");
throw std::invalid_argument("uri must be base64 encoded");
} else {
auto base64_data = parts[1];
auto decoded_data = base64_decode(base64_data);
@@ -1219,7 +1220,7 @@ json oaicompat_chat_params_parse(
json image_url = json_value(p, "image_url", json::object());
std::string url = json_value(image_url, "url", std::string());
handle_media(out_files, url, opt.media_path, true);
handle_media(out_files, url, opt.media_path);
p["type"] = "media_marker";
p["text"] = get_media_marker();
@@ -1234,7 +1235,7 @@ json oaicompat_chat_params_parse(
json input_audio = json_value(p, "input_audio", json::object());
std::string url = json_value(input_audio, "data",
json_value(input_audio, "url", std::string()));
handle_media(out_files, url, opt.media_path, false);
handle_media(out_files, url, opt.media_path);
p["type"] = "media_marker";
p["text"] = get_media_marker();
@@ -1248,7 +1249,7 @@ json oaicompat_chat_params_parse(
json input_video = json_value(p, "input_video", json::object());
std::string url = json_value(input_video, "data",
json_value(input_video, "url", std::string()));
handle_media(out_files, url, opt.media_path, false);
handle_media(out_files, url, opt.media_path);
p["type"] = "media_marker";
p["text"] = get_media_marker();
tools/server/server-context.cpp
+13 -2
View File
@@ -1493,11 +1493,22 @@ private:
auto caps = common_chat_templates_get_caps(chat_params.tmpls.get());
auto it = params_base.default_template_kwargs.find("preserve_reasoning");
bool supported = caps.at("supports_preserve_reasoning");
bool enabled = it != params_base.default_template_kwargs.end();
bool specified = params_base.preserve_reasoning_specified;
// note: the kwarg is enabled by default if not specified explicitly, so check the value
bool enabled = it != params_base.default_template_kwargs.end() && it->second == "true";
if (supported) {
SRV_TRC("preserve_reasoning kwarg: %s\n",
it == params_base.default_template_kwargs.end() ? "unset (template default)" : it->second.c_str());
} else {
SRV_TRC("%s", "preserve_reasoning kwarg: not supported by template\n");
}
if (supported && !specified) {
SRV_WRN("%s", "chat template supports preserving reasoning, it is enabled by default (may use more tokens, disable via --no-reasoning-preserve)\n");
}
if (supported && !enabled) {
SRV_INF("%s", "chat template supports preserving reasoning, consider enabling it via --reasoning-preserve\n");
}
if (!supported && enabled) {
if (!supported && specified && enabled) {
SRV_WRN("%s", "chat template does NOT support preserving reasoning, --reasoning-preserve has no effect\n");
}
}
tools/server/tests/unit/test_vision_api.py
+1
View File
@@ -71,6 +71,7 @@ def test_v1_models_supports_multimodal_capability():
("What is this:\n", "malformed", False, None),
("What is this:\n", "https://google.com/404", False, None), # non-existent image
("What is this:\n", "https://ggml.ai", False, None), # non-image data
("What is this:\n", "data:text/html;base64,aGVsbG8=", False, None), # unsupported data uri mime
# TODO @ngxson : test with multiple images, no images and with audio
]
)