Merge commit '1c3c9674de4d455f1e571bed808252af54932767' into concedo_experimental

# Conflicts:
#	.github/workflows/build-apple.yml
#	.github/workflows/build-vulkan.yml
#	.github/workflows/release.yml
#	docs/ops.md
#	docs/ops/Vulkan.csv
#	examples/gen-docs/gen-docs.cpp
#	ggml/CMakeLists.txt
#	ggml/src/ggml-opencl/ggml-opencl.cpp
#	ggml/src/ggml-sycl/concat.cpp
#	ggml/src/ggml-sycl/fattn-onednn.cpp
#	ggml/src/ggml-sycl/fattn.cpp
#	models/templates/deepseek-ai-DeepSeek-V4.jinja
#	scripts/sync-ggml.last
#	scripts/sync_vendor.py
#	src/CMakeLists.txt
#	src/llama-model-loader.cpp
#	tests/CMakeLists.txt
#	tests/test-arg-parser.cpp
#	tests/test-backend-sampler.cpp
#	tests/test-chat.cpp
#	tests/test-sampling.cpp
#	tools/server/README.md
This commit is contained in:
Concedo
2026-08-07 18:16:11 +08:00
51 changed files with 2686 additions and 780 deletions
+20 -5
View File
@@ -28,6 +28,7 @@
#include <algorithm>
#include <cinttypes>
#include <climits>
#include <cmath>
#include <cstdarg>
#include <filesystem>
#include <fstream>
@@ -2037,7 +2038,13 @@ common_params_context common_params_parser_init(common_params & params, llama_ex
{"--repeat-penalty"}, "N",
string_format("penalize repeat sequence of tokens (default: %.2f, 1.0 = disabled)", (double)params.sampling.penalty_repeat),
[](common_params & params, const std::string & value) {
params.sampling.penalty_repeat = std::stof(value);
const float penalty_repeat = std::stof(value);
if (!std::isfinite(penalty_repeat) ||
penalty_repeat <= 0.0f ||
!std::isfinite(1.0f/penalty_repeat)) {
throw std::runtime_error("error: repeat-penalty must be finite and greater than 0\n");
}
params.sampling.penalty_repeat = penalty_repeat;
params.sampling.user_sampling_config |= common_params_sampling_config::COMMON_PARAMS_SAMPLING_CONFIG_PENALTY_REPEAT;
}
).set_sampling());
@@ -2045,14 +2052,22 @@ common_params_context common_params_parser_init(common_params & params, llama_ex
{"--presence-penalty"}, "N",
string_format("repeat alpha presence penalty (default: %.2f, 0.0 = disabled)", (double)params.sampling.penalty_present),
[](common_params & params, const std::string & value) {
params.sampling.penalty_present = std::stof(value);
const float penalty_present = std::stof(value);
if (!std::isfinite(penalty_present)) {
throw std::runtime_error("error: presence-penalty must be finite\n");
}
params.sampling.penalty_present = penalty_present;
}
).set_sampling());
add_opt(common_arg(
{"--frequency-penalty"}, "N",
string_format("repeat alpha frequency penalty (default: %.2f, 0.0 = disabled)", (double)params.sampling.penalty_freq),
[](common_params & params, const std::string & value) {
params.sampling.penalty_freq = std::stof(value);
const float penalty_freq = std::stof(value);
if (!std::isfinite(penalty_freq)) {
throw std::runtime_error("error: frequency-penalty must be finite\n");
}
params.sampling.penalty_freq = penalty_freq;
}
).set_sampling());
add_opt(common_arg(
@@ -2568,7 +2583,7 @@ common_params_context common_params_parser_init(common_params & params, llama_ex
params.mtmd_batch_max_tokens = value;
}
).set_examples({LLAMA_EXAMPLE_SERVER}).set_env("LLAMA_ARG_MTMD_BATCH_MAX_TOKENS"));
if (llama_supports_rpc()) {
if (params.is_gen_docs || llama_supports_rpc()) {
add_opt(common_arg(
{"--rpc"}, "SERVERS",
"comma-separated list of RPC servers (host:port)",
@@ -3317,7 +3332,7 @@ common_params_context common_params_parser_init(common_params & params, llama_ex
{"--tools"}, "TOOL1,TOOL2,...",
"experimental: whether to enable built-in tools for AI agents - do not enable in untrusted environments (default: no tools)\n"
"specify \"all\" to enable all tools\n"
"available tools: read_file, file_glob_search, grep_search, exec_shell_command, write_file, edit_file, get_datetime\n"
"available tools: read_file, file_glob_search, grep_search, exec_shell_command, write_file, edit_file, get_datetime, get_info\n"
"note: for security reasons, this will limit --cors-origins to localhost by default",
[](common_params & params, const std::string & value) {
params.server_tools = parse_csv_row(value);