ggml 0.25.3

ggml/.pi/gg/SYSTEM.md

@@ -2,23 +2,28 @@ You are a coding agent. Here are some very important rules that you must follow:
 
 General:
 - Be very precise and concise when writing code, comments, explanations, etc.
+- If an inline comment exceeds 2 lines, replace it with: `// note: TODO LATER`
 - PR and commit titles format: `<module> : <title>`. Lookup recents for examples
 - Don't try to build or run the code unless you are explicitly asked to do so
 - Use the `gh` CLI tool when querying PRs, issues, or other GitHub resources
+- When [MODEL] is needed, first try to get it from the `PI_MODEL_NAME` env var before asking the user
 
 Coding:
 - When in doubt, always refer to the CONTRIBUTING.md file of the project
+- In `test-backend-ops.cpp`, do not mention specific backends (e.g. Metal, CUDA) in comments
 - When referencing issues or PRs in comments, use the format:
   - C/C++ code: `// ref: <url>`
   - Other (CMake, etc.): `# ref: <url>`
 
 Pull requests (PRs):
 - New branch names are prefixed with "gg/"
 - Before opening a pull request, ask the user to confirm the description
+- Don't explicitly wrap lines in the PR description (each paragraph and bullet is a single line)
 - When creating a pull request, look for the repository's PR template and follow it
 - For the AI usage disclosure section, write "YES. pi:llama.cpp/[MODEL]"
-- Ask the user to tell you what model was used and write it in place of [MODEL]
+- If `PI_MODEL_NAME` env var is not set, ask the user to tell you what model was used and write it in place of [MODEL]
 - Always create the pull requests in draft mode
+- Never reply to review comments or post comments on issues/PRs without explicit permission from the user
 
 Commits:
 - On every commit that you make, include a "Assisted-by: pi:llama.cpp/[MODEL]" tag

ggml/CMakeLists.txt

@@ -5,7 +5,7 @@ project("ggml" C CXX ASM)
 ### GGML Version
 set(GGML_VERSION_MAJOR 0)
 set(GGML_VERSION_MINOR 25)
-set(GGML_VERSION_PATCH 1)
+set(GGML_VERSION_PATCH 3)
 set(GGML_VERSION_BASE "${GGML_VERSION_MAJOR}.${GGML_VERSION_MINOR}.${GGML_VERSION_PATCH}")
 
 list(APPEND CMAKE_MODULE_PATH "${CMAKE_CURRENT_SOURCE_DIR}/cmake/")

ggml/UPSTREAM

@@ -1,5 +1,5 @@
 repo: git@github.com:ggml-org/ggml.git
-sha:  e565a8f4ce2e462c4973a24c51098dd3c81c0256
+sha:  353b63b439f27ab2cc19dac97ab1681ba6d2d084
 patches:
   patches/ggml/0001-fix-threadpool-oversubscription.patch
 

ggml/requirements.txt

@@ -1,12 +1,11 @@
 accelerate==0.19.0
-numpy~=1.26.4; python_version < "3.13"
-numpy~=2.1.0; python_version >= "3.13"
+numpy~=2.2.6
 sentencepiece>=0.1.98,<0.3.0
-torchvision~=0.21.0
+torchvision~=0.26.0
 transformers==5.5.1
 gguf>=0.1.0
 keras==3.10.0
-tensorflow==2.20.0
+tensorflow==2.20.0; python_version < "3.14"
 
 --extra-index-url https://download.pytorch.org/whl/cpu
-torch~=2.6.0
+torch~=2.11.0

ggml/scripts/make-release-desc.sh

@@ -35,6 +35,15 @@ if ! git fetch --tags origin 2>/dev/null; then
     echo "Warning: could not fetch tags from origin (local run?)"
 fi
 
+# Canonical https URL of this repository (from the origin remote), used to link the previous release.
+# Left empty on local runs without an origin remote.
+if ORIGIN_URL="$(git remote get-url origin 2>/dev/null)"; then
+    REPO_URL="https://$(printf '%s' "${ORIGIN_URL}" \
+        | sed -E -e 's#^git@([^:]+):#https://\1/#' -e 's#^https?://##' -e 's#\.git$##')"
+else
+    REPO_URL=""
+fi
+
 # Release commit: the commit <version> points at when the tag exists, HEAD otherwise.
 if ! RELEASE_COMMIT="$(git rev-parse -q --verify "refs/tags/${VERSION}^{commit}" 2>/dev/null)"; then
     RELEASE_COMMIT="$(git rev-parse HEAD)"
@@ -49,7 +58,11 @@ PREV="$( { git tag --list; echo "${VERSION}"; } \
 
 if [[ -n "${PREV}" ]]; then
     CHANGELOG="$(git log --oneline "${PREV}..${RELEASE_COMMIT}")"
-    CHANGELOG_TITLE="Changelog since ${PREV}"
+    if [[ -n "${REPO_URL}" ]]; then
+        CHANGELOG_TITLE="Changelog since [${PREV}](${REPO_URL}/releases/tag/${PREV})"
+    else
+        CHANGELOG_TITLE="Changelog since ${PREV}"
+    fi
 else
     CHANGELOG="(no previous release tag found)"
     CHANGELOG_TITLE="Changelog"

ggml/scripts/sync-llama.last

@@ -1 +1 @@
-66fba63af1f4161052c33024d150cac31f46ff37
+6b790a9c291b5d7af3312bbf9f0c558aa023b13e