sync : ggml

.devops/main-intel.Dockerfile

@@ -1,6 +1,6 @@
-ARG ONEAPI_VERSION=2025.1.1-0-devel-ubuntu24.04
+ARG ONEAPI_VERSION=2025.3.3-0-devel-ubuntu24.04
 
-FROM intel/oneapi-basekit:$ONEAPI_VERSION AS build
+FROM intel/deep-learning-essentials:$ONEAPI_VERSION AS build
 WORKDIR /app
 
 RUN apt-get update && \
@@ -16,7 +16,7 @@ RUN if [ "${GGML_SYCL_F16}" = "ON" ]; then \
     fi && \
     make base.en CMAKE_ARGS="-DGGML_SYCL=1 -DCMAKE_C_COMPILER=icx -DCMAKE_CXX_COMPILER=icpx ${OPT_SYCL_F16}"
 
-FROM intel/oneapi-basekit:$ONEAPI_VERSION AS runtime
+FROM intel/deep-learning-essentials:$ONEAPI_VERSION AS runtime
 WORKDIR /app
 
 RUN apt-get update && \

.devops/main-vulkan.Dockerfile

@@ -2,7 +2,7 @@ FROM ubuntu:24.04 AS build
 WORKDIR /app
 
 RUN apt-get update && \
-  apt-get install -y build-essential wget cmake git libvulkan-dev glslc \
+  apt-get install -y build-essential wget cmake git libvulkan-dev spirv-headers glslc \
   && rm -rf /var/lib/apt/lists/* /var/cache/apt/archives/*
 
 COPY .. .

examples/talk-llama/llama-arch.cpp

@@ -232,6 +232,7 @@ static const std::map<llm_kv, const char *> LLM_KV_NAMES = {
     { LLM_KV_ATTENTION_SLIDING_WINDOW_PATTERN,       "%s.attention.sliding_window_pattern"       },
     { LLM_KV_ATTENTION_SCALE,                        "%s.attention.scale"                        },
     { LLM_KV_ATTENTION_OUTPUT_SCALE,                 "%s.attention.output_scale"                 },
+    { LLM_KV_ATTENTION_VALUE_SCALE,                  "%s.attention.value_scale"                  },
     { LLM_KV_ATTENTION_TEMPERATURE_LENGTH,           "%s.attention.temperature_length"           },
     { LLM_KV_ATTENTION_TEMPERATURE_SCALE,            "%s.attention.temperature_scale"            },
     { LLM_KV_ATTENTION_KEY_LENGTH_MLA,               "%s.attention.key_length_mla"               },

examples/talk-llama/llama-arch.h

@@ -236,6 +236,7 @@ enum llm_kv {
     LLM_KV_ATTENTION_SLIDING_WINDOW_PATTERN,
     LLM_KV_ATTENTION_SCALE,
     LLM_KV_ATTENTION_OUTPUT_SCALE,
+    LLM_KV_ATTENTION_VALUE_SCALE,
     LLM_KV_ATTENTION_TEMPERATURE_LENGTH,
     LLM_KV_ATTENTION_TEMPERATURE_SCALE,
     LLM_KV_ATTENTION_KEY_LENGTH_MLA,