Merge branch 'upstream' into concedo_experimental

# Conflicts:
#	docs/backend/SYCL.md
#	docs/backend/snapdragon/developer.md
#	examples/convert-llama2c-to-ggml/convert-llama2c-to-ggml.cpp
#	examples/sycl/run-llama2.sh
#	examples/sycl/start-svr.sh
#	examples/sycl/test.sh
#	examples/sycl/win-run-llama2.bat
#	examples/sycl/win-start-svr.bat
#	examples/sycl/win-test.bat
#	ggml/CMakeLists.txt
#	ggml/src/ggml-et/et-kernels/src/ssm_scan_f32.c
#	ggml/src/ggml-et/ggml-et-ops.cpp
#	ggml/src/ggml-et/ggml-et-ops.h
#	ggml/src/ggml-sycl/ssm_scan.cpp
#	ggml/src/ggml-webgpu/ggml-webgpu.cpp
#	ggml/src/ggml-webgpu/wgsl-shaders/ssm_scan.wgsl
#	scripts/bench-models.sh
#	scripts/snapdragon/adb/run-bench.sh
#	scripts/snapdragon/adb/run-cli.sh
#	scripts/snapdragon/adb/run-completion.sh
#	scripts/snapdragon/adb/run-mtmd.sh
#	scripts/snapdragon/windows/run-bench.ps1
#	scripts/snapdragon/windows/run-cli.ps1
#	scripts/snapdragon/windows/run-completion.ps1
#	scripts/snapdragon/windows/run-mtmd.ps1
#	scripts/sync-ggml.last
#	scripts/sync_vendor.py
#	tests/CMakeLists.txt
#	tests/test-backend-ops.cpp
#	tests/test-chat.cpp
#	tests/test-jinja.cpp
#	tests/test-llama-archs.cpp
#	tools/cli/README.md
#	tools/completion/README.md
#	tools/llama-bench/README.md
#	tools/server/README.md
This commit is contained in:
Concedo
2026-08-15 22:50:04 +08:00
60 changed files with 1781 additions and 447 deletions
+41 -8
View File
@@ -20,7 +20,7 @@ namespace jinja {
using caps_json_fn = std::function<json()>;
using caps_ctx_fn = std::function<void(context &)>;
using caps_analyze_fn = std::function<void(bool, value &, value &, const std::string &)>;
using caps_analyze_fn = std::function<void(context &, bool, value &, value &, const std::string &)>;
void caps_apply_preserve_reasoning(jinja::context & ctx, bool enabled) {
ctx.set_val("preserve_thinking", mk_val<value_bool>(enabled));
@@ -29,6 +29,12 @@ void caps_apply_preserve_reasoning(jinja::context & ctx, bool enabled) {
ctx.set_val("drop_thinking", mk_val<value_bool>(!enabled));
}
void caps_apply_reasoning_effort(jinja::context & ctx, const std::string & effort) {
value var = mk_val<value_string>(effort); // bind to the same value for stats
ctx.set_val("reasoning_effort", var);
ctx.set_val("reasoning_strength", var);
}
static void caps_try_execute(jinja::program & prog,
const caps_json_fn & messages_fn,
const caps_ctx_fn & ctx_fn,
@@ -65,7 +71,7 @@ static void caps_try_execute(jinja::program & prog,
// ignore exceptions during capability analysis
}
analyze_fn(success, messages, tools, result);
analyze_fn(ctx, success, messages, tools, result);
}
// for debugging only
@@ -90,6 +96,7 @@ std::map<std::string, bool> caps::to_map() const {
{"supports_parallel_tool_calls", supports_parallel_tool_calls},
{"supports_system_role", supports_system_role},
{"supports_preserve_reasoning", supports_preserve_reasoning},
{"supports_reasoning_effort", supports_reasoning_effort},
{"supports_object_arguments", supports_object_arguments},
};
}
@@ -127,7 +134,7 @@ caps caps_get(jinja::program & prog) {
},
nullptr, // ctx_fn
nullptr, // tools_fn
[&](bool success, value & messages, value &, const std::string &) {
[&](context &, bool success, value & messages, value &, const std::string &) {
auto & content = messages->at(0)->at("content");
caps_print_stats(content, "messages[0].content");
if (has_op(content, "selectattr") || has_op(content, "array_access")) {
@@ -161,7 +168,7 @@ caps caps_get(jinja::program & prog) {
},
nullptr, // ctx_fn
nullptr, // tools_fn
[&](bool, value & messages, value &, const std::string &) {
[&](context &, bool, value & messages, value &, const std::string &) {
auto & content = messages->at(0)->at("content");
caps_print_stats(content, "messages[0].content");
if (!content->stats.used) {
@@ -237,7 +244,7 @@ caps caps_get(jinja::program & prog) {
},
});
},
[&](bool success, value & messages, value & tools, const std::string &) {
[&](context &, bool success, value & messages, value & tools, const std::string &) {
if (!success) {
return; // Nothing can be inferred
}
@@ -330,7 +337,7 @@ caps caps_get(jinja::program & prog) {
},
});
},
[&](bool success, value & messages, value & tools, const std::string &) {
[&](context &, bool success, value & messages, value & tools, const std::string &) {
if (!success) {
result.supports_tool_calls = false;
result.supports_tools = false;
@@ -432,7 +439,7 @@ caps caps_get(jinja::program & prog) {
},
});
},
[&](bool success, value & messages, value &, const std::string &) {
[&](context &, bool success, value & messages, value &, const std::string &) {
if (!success) {
result.supports_parallel_tool_calls = false;
return;
@@ -489,7 +496,7 @@ caps caps_get(jinja::program & prog) {
caps_apply_preserve_reasoning(ctx, true);
},
nullptr, // tools_fn
[&](bool, value &, value &, const std::string & output) {
[&](context &, bool, value &, value &, const std::string & output) {
// note: we cannot use stats here because the reasoning_content may be used for "if" condition test, but not actually outputted in the final result
if (output.find(reasoning_placeholder) != std::string::npos) {
result.supports_preserve_reasoning = true;
@@ -497,6 +504,32 @@ caps caps_get(jinja::program & prog) {
}
);
JJ_DEBUG("%s\n", ">>> Running capability check: reasoning effort");
// case: reasoning effort level
caps_try_execute(
prog,
[&]() {
// messages
return json::array({
{
{"role", "user"},
{"content", "User message"}
},
});
},
[&](context & ctx) {
ctx.set_val("enable_thinking", mk_val<value_bool>(true));
caps_apply_reasoning_effort(ctx, "low");
},
nullptr, // tools_fn
[&](context & ctx, bool, value &, value &, const std::string &) {
value effort = ctx.get_val("reasoning_effort");
caps_print_stats(effort, "reasoning_effort");
result.supports_reasoning_effort = effort->stats.used;
}
);
JJ_DEBUG("%s\n", result.to_string().c_str());
return result;