{"entries":[{"name":"ai-issues.yml","mode":"0100644","size":2904,"lastCommit":{"hash":"e32d2438496b627d04846874aeb48978999ee4d3","oid":"e32d2438496b627d04846874aeb48978999ee4d3","parents":["c44a932cf495dc1cf73020f62acce6233a0a95f3"],"author":{"name":"Georgi Gerganov","email":"ggerganov@gmail.com","when":"2026-03-23T13:21:41+02:00"},"committer":{"name":"GitHub","email":"noreply@github.com","when":"2026-03-23T13:21:41+02:00"},"message":"ai : update gh permissions (#20895)","tree":"5b4e9481f93ae2d66b5de1b077019c164b2b4c11"}},{"name":"bench.yml.disabled","mode":"0100644","size":10473,"lastCommit":{"hash":"1d36b3670b285e69e58b9d687c770a2a0a192194","oid":"1d36b3670b285e69e58b9d687c770a2a0a192194","parents":["b34443923cad751483cc53af2e680d595daadce7"],"author":{"name":"Diego Devesa","email":"slarengh@gmail.com","when":"2025-05-02T20:27:13+02:00"},"committer":{"name":"GitHub","email":"noreply@github.com","when":"2025-05-02T20:27:13+02:00"},"message":"llama : move end-user examples to tools directory (#13249)\n\n* llama : move end-user examples to tools directory\n\n---------\n\nCo-authored-by: Xuan Son Nguyen ","tree":"a510b2a686d850dde8a4da94b74fc718df8d8b9c"}},{"name":"build-3rd-party.yml","mode":"0100644","size":1213,"lastCommit":{"hash":"6b4e4bd582e457a47ee7cd498058495f47fee228","oid":"6b4e4bd582e457a47ee7cd498058495f47fee228","parents":["9f0e4b14d23aeedb106219c67a64cd7bbd19b63e"],"author":{"name":"Georgi Gerganov","email":"ggerganov@gmail.com","when":"2026-05-27T14:52:47+03:00"},"committer":{"name":"GitHub","email":"noreply@github.com","when":"2026-05-27T14:52:47+03:00"},"message":"common : fix env names to all have LLAMA_ARG_ prefix (#23778)","tree":"23e83fbcf9cb68c4e84c8fc4d2190b9fb5a33de1"}},{"name":"build-and-test-snapdragon.yml","mode":"0100644","size":4856,"lastCommit":{"hash":"4bead4e30d3ee3f6baef7da2833210374c300d33","oid":"4bead4e30d3ee3f6baef7da2833210374c300d33","parents":["302e2c2652ba42992f6df581a191a076151fce94"],"author":{"name":"Max Krasnyansky","email":"maxk@qti.qualcomm.com","when":"2026-05-25T10:57:43-07:00"},"committer":{"name":"GitHub","email":"noreply@github.com","when":"2026-05-25T10:57:43-07:00"},"message":"snapdragon: bump toolchain docker to v0.7 to fix ui build issues (#23680)","tree":"ab33aa8076f80f062889cbd050b993bb993cc3bd"}},{"name":"build-android.yml","mode":"0100644","size":4295,"lastCommit":{"hash":"7dffb158de30ebb8ef9d64f33c6b0b2d7c1e6313","oid":"7dffb158de30ebb8ef9d64f33c6b0b2d7c1e6313","parents":["6ac5eb087009a762a3f61876f7dc152bc8600bf4"],"author":{"name":"Pasha Khosravi","email":"khosravipasha@users.noreply.github.com","when":"2026-09-15T02:04:13-07:00"},"committer":{"name":"GitHub","email":"noreply@github.com","when":"2026-09-15T02:04:13-07:00"},"message":"ci: install only platform-tools with setup-android (legacy 'tools' package is gone) (#175)","tree":"b21f8b1354ee48e941434e441efa5db658c2f1cb"}},{"name":"build-apple.yml","mode":"0100644","size":7715,"lastCommit":{"hash":"985b14912bc9a3d8f27c1ac569e46f4092123722","oid":"985b14912bc9a3d8f27c1ac569e46f4092123722","parents":["6036c635e2d1e9aeca11d01a84188c9db9ea1e2a"],"author":{"name":"Georgi Gerganov","email":"ggerganov@gmail.com","when":"2026-08-24T10:49:20+03:00"},"committer":{"name":"GitHub","email":"noreply@github.com","when":"2026-08-24T10:49:20+03:00"},"message":"ci : apply ccache-clear with older/min/dry-run to all ccache jobs (#27602)\n\n* ci : apply ccache-clear with older/min/dry-run to all ccache jobs\n\nAssisted-by: llama.cpp:DeepSeek-v4-Flash-0731\n\n* ci : install gh in ccache-clear if missing (container jobs)\n\nThe ccache-clear action relies on the gh CLI, which is not present in\ncontainer-based jobs. Install it on demand so those jobs can clear caches.\n\nAssisted-by: llama.cpp:DeepSeek-v4-Flash-0731\n\n* ci : install gh via apt repo in ccache-clear\n\nThe install.sh script used previously is no longer served (404). Switch to\nthe official GitHub CLI apt repository, which is still available.\n\nAssisted-by: llama.cpp:DeepSeek-v4-Flash-0731\n\n* ci : pass --repo to gh cache commands in ccache-clear\n\nIn container jobs gh cannot auto-detect the repository from git, so\ngh cache list/delete fail with 'failed to run git: not a git repository'.\nPass the repository explicitly via --repo using GITHUB_REPOSITORY.\n\nAssisted-by: llama.cpp:DeepSeek-v4-Flash-0731\n\n* ci : drop -new suffix from vulkan ccache key\n\nThe -new suffix was only needed to force a fresh cache. With\nccache-clear now evicting stale caches, the original key can be used\nagain. The old ccache-vulkan-ubuntu-24.04-arm-new entries still match\nthe ccache-clear key prefix and are cleaned up automatically.\n\nAssisted-by: llama.cpp:DeepSeek-v4-Flash-0731\n\n* ci : fix ccache-clear date parsing on macOS (BSD date)\n\nmacOS ships BSD date, which has no -d option. The older cutoff check\nwas silently disabled there: 'date: illegal option -- d' errors in the\nlog and the loop was only stopped by the min limit, risking deletion\nof caches not older than the cutoff (e.g. saved by a concurrent job).\n\nParse the ISO-8601 timestamps with GNU date when available and fall\nback to BSD date otherwise (TZ=UTC, fractional seconds dropped).\n\nAssisted-by: llama.cpp:DeepSeek-v4-Flash-0731\n\n* ci : extract ccache-clear logic into scripts/ccache-clear.sh\n\nThe composite action now consists of a dedicated step that installs the\nGitHub CLI when missing (e.g. in container jobs) and a thin step that\ncalls the new script. The script follows the make-release-checks.sh\nconventions (usage/env header, set -euo pipefail, CLI flags) and only\nchecks that gh is available. The action inputs are unchanged, so the\nworkflow steps are untouched.\n\nAssisted-by: llama.cpp:DeepSeek-v4-Flash-0731\n\n* ci : remove unused apple ccaches","tree":"0f6df18eb929856daca94b05ada439528798dd6a"}},{"name":"build-cache.yml","mode":"0100644","size":3570,"lastCommit":{"hash":"9d77fa17254e1dee4b9e92504c91611a60b1359f","oid":"9d77fa17254e1dee4b9e92504c91611a60b1359f","parents":["c0296022f361beaab4c6ecf5def391d2fc598dae"],"author":{"name":"Zijun Yu","email":"zijun.yu@intel.com","when":"2026-08-18T18:02:22+08:00"},"committer":{"name":"GitHub","email":"noreply@github.com","when":"2026-08-18T12:02:22+02:00"},"message":"ci : Update OpenVINO to 2026.3, skip nemotron-h rollback test (#27292)\n\n* update to ov-2026.3, update device drivers\n\n* ci: skip nemotron-h rollback test on OpenVINO\n\nThe OpenVINO backend does not support SSM_SCAN, so the Nemotron-H recurrent state rollback graph is split and cannot preserve the recurrent cache output shape. Keep the test enabled for other backends and retain the qwen35 OpenVINO rollback coverage.\n\n---------\n\nCo-authored-by: ravi9 ","tree":"4209b7a0a8180a2c50a1c605b8b913e52798a9cd"}},{"name":"build-cann.yml","mode":"0100644","size":3309,"lastCommit":{"hash":"6b4e4bd582e457a47ee7cd498058495f47fee228","oid":"6b4e4bd582e457a47ee7cd498058495f47fee228","parents":["9f0e4b14d23aeedb106219c67a64cd7bbd19b63e"],"author":{"name":"Georgi Gerganov","email":"ggerganov@gmail.com","when":"2026-05-27T14:52:47+03:00"},"committer":{"name":"GitHub","email":"noreply@github.com","when":"2026-05-27T14:52:47+03:00"},"message":"common : fix env names to all have LLAMA_ARG_ prefix (#23778)","tree":"23e83fbcf9cb68c4e84c8fc4d2190b9fb5a33de1"}},{"name":"build-cmake-pkg.yml","mode":"0100644","size":2024,"lastCommit":{"hash":"0e1d9185c5fe82e905d1f5ae6b2e5dcd607a8dfd","oid":"0e1d9185c5fe82e905d1f5ae6b2e5dcd607a8dfd","parents":["a30273376ef669023334fc20ad02ae4ed8196a65"],"author":{"name":"Eve","email":"139727413+netrunnereve@users.noreply.github.com","when":"2026-08-20T20:01:32Z"},"committer":{"name":"GitHub","email":"noreply@github.com","when":"2026-08-20T20:01:32Z"},"message":"ci: use shell script to check cmake pkg (#27414)\n\n* use regular script to build cmake pkg\n\n* use old grep without perl","tree":"63deaef603952cf031ff297648bbf79686c10b73"}},{"name":"build-cpu.yml","mode":"0100644","size":8298,"lastCommit":{"hash":"985b14912bc9a3d8f27c1ac569e46f4092123722","oid":"985b14912bc9a3d8f27c1ac569e46f4092123722","parents":["6036c635e2d1e9aeca11d01a84188c9db9ea1e2a"],"author":{"name":"Georgi Gerganov","email":"ggerganov@gmail.com","when":"2026-08-24T10:49:20+03:00"},"committer":{"name":"GitHub","email":"noreply@github.com","when":"2026-08-24T10:49:20+03:00"},"message":"ci : apply ccache-clear with older/min/dry-run to all ccache jobs (#27602)\n\n* ci : apply ccache-clear with older/min/dry-run to all ccache jobs\n\nAssisted-by: llama.cpp:DeepSeek-v4-Flash-0731\n\n* ci : install gh in ccache-clear if missing (container jobs)\n\nThe ccache-clear action relies on the gh CLI, which is not present in\ncontainer-based jobs. Install it on demand so those jobs can clear caches.\n\nAssisted-by: llama.cpp:DeepSeek-v4-Flash-0731\n\n* ci : install gh via apt repo in ccache-clear\n\nThe install.sh script used previously is no longer served (404). Switch to\nthe official GitHub CLI apt repository, which is still available.\n\nAssisted-by: llama.cpp:DeepSeek-v4-Flash-0731\n\n* ci : pass --repo to gh cache commands in ccache-clear\n\nIn container jobs gh cannot auto-detect the repository from git, so\ngh cache list/delete fail with 'failed to run git: not a git repository'.\nPass the repository explicitly via --repo using GITHUB_REPOSITORY.\n\nAssisted-by: llama.cpp:DeepSeek-v4-Flash-0731\n\n* ci : drop -new suffix from vulkan ccache key\n\nThe -new suffix was only needed to force a fresh cache. With\nccache-clear now evicting stale caches, the original key can be used\nagain. The old ccache-vulkan-ubuntu-24.04-arm-new entries still match\nthe ccache-clear key prefix and are cleaned up automatically.\n\nAssisted-by: llama.cpp:DeepSeek-v4-Flash-0731\n\n* ci : fix ccache-clear date parsing on macOS (BSD date)\n\nmacOS ships BSD date, which has no -d option. The older cutoff check\nwas silently disabled there: 'date: illegal option -- d' errors in the\nlog and the loop was only stopped by the min limit, risking deletion\nof caches not older than the cutoff (e.g. saved by a concurrent job).\n\nParse the ISO-8601 timestamps with GNU date when available and fall\nback to BSD date otherwise (TZ=UTC, fractional seconds dropped).\n\nAssisted-by: llama.cpp:DeepSeek-v4-Flash-0731\n\n* ci : extract ccache-clear logic into scripts/ccache-clear.sh\n\nThe composite action now consists of a dedicated step that installs the\nGitHub CLI when missing (e.g. in container jobs) and a thin step that\ncalls the new script. The script follows the make-release-checks.sh\nconventions (usage/env header, set -euo pipefail, CLI flags) and only\nchecks that gh is available. The action inputs are unchanged, so the\nworkflow steps are untouched.\n\nAssisted-by: llama.cpp:DeepSeek-v4-Flash-0731\n\n* ci : remove unused apple ccaches","tree":"0f6df18eb929856daca94b05ada439528798dd6a"}},{"name":"build-cross.yml","mode":"0100644","size":13660,"lastCommit":{"hash":"3dc7684f3939d820742be0a4f43d6b2bd3035c01","oid":"3dc7684f3939d820742be0a4f43d6b2bd3035c01","parents":["dbe9c0c8ce65354c372f5d4ab507e5424a755e9f"],"author":{"name":"Georgi Gerganov","email":"ggerganov@gmail.com","when":"2026-05-26T15:21:21+03:00"},"committer":{"name":"GitHub","email":"noreply@github.com","when":"2026-05-26T15:21:21+03:00"},"message":"ci : reduce (disable SYCL and CANN builds/releases) (#23705)\n\n* ci : reduce\n\n[no ci]\n\n* cont : disable sycl, cann + rename caches\n\n[no ci]\n\n* cont : cann\n\n[no ci]","tree":"16cb488500daa534fbb3d24993837fb1d0ebc00c"}},{"name":"build-cuda-ubuntu.yml","mode":"0100644","size":4517,"lastCommit":{"hash":"985b14912bc9a3d8f27c1ac569e46f4092123722","oid":"985b14912bc9a3d8f27c1ac569e46f4092123722","parents":["6036c635e2d1e9aeca11d01a84188c9db9ea1e2a"],"author":{"name":"Georgi Gerganov","email":"ggerganov@gmail.com","when":"2026-08-24T10:49:20+03:00"},"committer":{"name":"GitHub","email":"noreply@github.com","when":"2026-08-24T10:49:20+03:00"},"message":"ci : apply ccache-clear with older/min/dry-run to all ccache jobs (#27602)\n\n* ci : apply ccache-clear with older/min/dry-run to all ccache jobs\n\nAssisted-by: llama.cpp:DeepSeek-v4-Flash-0731\n\n* ci : install gh in ccache-clear if missing (container jobs)\n\nThe ccache-clear action relies on the gh CLI, which is not present in\ncontainer-based jobs. Install it on demand so those jobs can clear caches.\n\nAssisted-by: llama.cpp:DeepSeek-v4-Flash-0731\n\n* ci : install gh via apt repo in ccache-clear\n\nThe install.sh script used previously is no longer served (404). Switch to\nthe official GitHub CLI apt repository, which is still available.\n\nAssisted-by: llama.cpp:DeepSeek-v4-Flash-0731\n\n* ci : pass --repo to gh cache commands in ccache-clear\n\nIn container jobs gh cannot auto-detect the repository from git, so\ngh cache list/delete fail with 'failed to run git: not a git repository'.\nPass the repository explicitly via --repo using GITHUB_REPOSITORY.\n\nAssisted-by: llama.cpp:DeepSeek-v4-Flash-0731\n\n* ci : drop -new suffix from vulkan ccache key\n\nThe -new suffix was only needed to force a fresh cache. With\nccache-clear now evicting stale caches, the original key can be used\nagain. The old ccache-vulkan-ubuntu-24.04-arm-new entries still match\nthe ccache-clear key prefix and are cleaned up automatically.\n\nAssisted-by: llama.cpp:DeepSeek-v4-Flash-0731\n\n* ci : fix ccache-clear date parsing on macOS (BSD date)\n\nmacOS ships BSD date, which has no -d option. The older cutoff check\nwas silently disabled there: 'date: illegal option -- d' errors in the\nlog and the loop was only stopped by the min limit, risking deletion\nof caches not older than the cutoff (e.g. saved by a concurrent job).\n\nParse the ISO-8601 timestamps with GNU date when available and fall\nback to BSD date otherwise (TZ=UTC, fractional seconds dropped).\n\nAssisted-by: llama.cpp:DeepSeek-v4-Flash-0731\n\n* ci : extract ccache-clear logic into scripts/ccache-clear.sh\n\nThe composite action now consists of a dedicated step that installs the\nGitHub CLI when missing (e.g. in container jobs) and a thin step that\ncalls the new script. The script follows the make-release-checks.sh\nconventions (usage/env header, set -euo pipefail, CLI flags) and only\nchecks that gh is available. The action inputs are unchanged, so the\nworkflow steps are untouched.\n\nAssisted-by: llama.cpp:DeepSeek-v4-Flash-0731\n\n* ci : remove unused apple ccaches","tree":"0f6df18eb929856daca94b05ada439528798dd6a"}},{"name":"build-cuda-windows.yml","mode":"0100644","size":6357,"lastCommit":{"hash":"c0296022f361beaab4c6ecf5def391d2fc598dae","oid":"c0296022f361beaab4c6ecf5def391d2fc598dae","parents":["7acdbb1f191d869bad8c5da9d4a2121defa340af"],"author":{"name":"shivamkumard-ctrl","email":"shivamkumard@nvidia.com","when":"2026-08-18T15:25:45+05:30"},"committer":{"name":"GitHub","email":"noreply@github.com","when":"2026-08-18T11:55:45+02:00"},"message":"ci: add Windows ARM64 CUDA support to the manual workflow (#27300)\n\n- Add a CUDA 13.4 ARM64 matrix entry.\n- Build only ggml-cuda for x64 and ARM64.","tree":"512134922aa0629b473e483382d77836f05aaa5b"}},{"name":"build-ibm.yml","mode":"0100644","size":4242,"lastCommit":{"hash":"6b4e4bd582e457a47ee7cd498058495f47fee228","oid":"6b4e4bd582e457a47ee7cd498058495f47fee228","parents":["9f0e4b14d23aeedb106219c67a64cd7bbd19b63e"],"author":{"name":"Georgi Gerganov","email":"ggerganov@gmail.com","when":"2026-05-27T14:52:47+03:00"},"committer":{"name":"GitHub","email":"noreply@github.com","when":"2026-05-27T14:52:47+03:00"},"message":"common : fix env names to all have LLAMA_ARG_ prefix (#23778)","tree":"23e83fbcf9cb68c4e84c8fc4d2190b9fb5a33de1"}},{"name":"build-msys.yml","mode":"0100644","size":1976,"lastCommit":{"hash":"46fa662b1f4cd3f00d774512cd50044b6d17bc2c","oid":"46fa662b1f4cd3f00d774512cd50044b6d17bc2c","parents":["7fe2ae45ab644c5b9b5740dff5068442f64fabce"],"author":{"name":"Daniel Bevenius","email":"daniel.bevenius@gmail.com","when":"2026-06-05T07:57:36+02:00"},"committer":{"name":"GitHub","email":"noreply@github.com","when":"2026-06-05T07:57:36+02:00"},"message":"ci : build-msys job slimming [no ci] (#24157)\n\nThis PR attempts to slim down the dependencies for build-msys jobs\nmaking the same changes that we applied in whisper.cpp to reduce the\nsize of the github actions cache, and should also improve the run time\ndue to fewer dependencies that need to be installed.\n\nI realize this is a scheduled job but I think it would still make sense\nto apply these changes.\n\nRefs: https://github.com/ggml-org/whisper.cpp/pull/3858","tree":"98a8caf5e19c00f42437ea16bfce0ac8dae3c3c3"}},{"name":"build-opencl.yml","mode":"0100644","size":2774,"lastCommit":{"hash":"985b14912bc9a3d8f27c1ac569e46f4092123722","oid":"985b14912bc9a3d8f27c1ac569e46f4092123722","parents":["6036c635e2d1e9aeca11d01a84188c9db9ea1e2a"],"author":{"name":"Georgi Gerganov","email":"ggerganov@gmail.com","when":"2026-08-24T10:49:20+03:00"},"committer":{"name":"GitHub","email":"noreply@github.com","when":"2026-08-24T10:49:20+03:00"},"message":"ci : apply ccache-clear with older/min/dry-run to all ccache jobs (#27602)\n\n* ci : apply ccache-clear with older/min/dry-run to all ccache jobs\n\nAssisted-by: llama.cpp:DeepSeek-v4-Flash-0731\n\n* ci : install gh in ccache-clear if missing (container jobs)\n\nThe ccache-clear action relies on the gh CLI, which is not present in\ncontainer-based jobs. Install it on demand so those jobs can clear caches.\n\nAssisted-by: llama.cpp:DeepSeek-v4-Flash-0731\n\n* ci : install gh via apt repo in ccache-clear\n\nThe install.sh script used previously is no longer served (404). Switch to\nthe official GitHub CLI apt repository, which is still available.\n\nAssisted-by: llama.cpp:DeepSeek-v4-Flash-0731\n\n* ci : pass --repo to gh cache commands in ccache-clear\n\nIn container jobs gh cannot auto-detect the repository from git, so\ngh cache list/delete fail with 'failed to run git: not a git repository'.\nPass the repository explicitly via --repo using GITHUB_REPOSITORY.\n\nAssisted-by: llama.cpp:DeepSeek-v4-Flash-0731\n\n* ci : drop -new suffix from vulkan ccache key\n\nThe -new suffix was only needed to force a fresh cache. With\nccache-clear now evicting stale caches, the original key can be used\nagain. The old ccache-vulkan-ubuntu-24.04-arm-new entries still match\nthe ccache-clear key prefix and are cleaned up automatically.\n\nAssisted-by: llama.cpp:DeepSeek-v4-Flash-0731\n\n* ci : fix ccache-clear date parsing on macOS (BSD date)\n\nmacOS ships BSD date, which has no -d option. The older cutoff check\nwas silently disabled there: 'date: illegal option -- d' errors in the\nlog and the loop was only stopped by the min limit, risking deletion\nof caches not older than the cutoff (e.g. saved by a concurrent job).\n\nParse the ISO-8601 timestamps with GNU date when available and fall\nback to BSD date otherwise (TZ=UTC, fractional seconds dropped).\n\nAssisted-by: llama.cpp:DeepSeek-v4-Flash-0731\n\n* ci : extract ccache-clear logic into scripts/ccache-clear.sh\n\nThe composite action now consists of a dedicated step that installs the\nGitHub CLI when missing (e.g. in container jobs) and a thin step that\ncalls the new script. The script follows the make-release-checks.sh\nconventions (usage/env header, set -euo pipefail, CLI flags) and only\nchecks that gh is available. The action inputs are unchanged, so the\nworkflow steps are untouched.\n\nAssisted-by: llama.cpp:DeepSeek-v4-Flash-0731\n\n* ci : remove unused apple ccaches","tree":"0f6df18eb929856daca94b05ada439528798dd6a"}},{"name":"build-openvino.yml","mode":"0100644","size":5803,"lastCommit":{"hash":"985b14912bc9a3d8f27c1ac569e46f4092123722","oid":"985b14912bc9a3d8f27c1ac569e46f4092123722","parents":["6036c635e2d1e9aeca11d01a84188c9db9ea1e2a"],"author":{"name":"Georgi Gerganov","email":"ggerganov@gmail.com","when":"2026-08-24T10:49:20+03:00"},"committer":{"name":"GitHub","email":"noreply@github.com","when":"2026-08-24T10:49:20+03:00"},"message":"ci : apply ccache-clear with older/min/dry-run to all ccache jobs (#27602)\n\n* ci : apply ccache-clear with older/min/dry-run to all ccache jobs\n\nAssisted-by: llama.cpp:DeepSeek-v4-Flash-0731\n\n* ci : install gh in ccache-clear if missing (container jobs)\n\nThe ccache-clear action relies on the gh CLI, which is not present in\ncontainer-based jobs. Install it on demand so those jobs can clear caches.\n\nAssisted-by: llama.cpp:DeepSeek-v4-Flash-0731\n\n* ci : install gh via apt repo in ccache-clear\n\nThe install.sh script used previously is no longer served (404). Switch to\nthe official GitHub CLI apt repository, which is still available.\n\nAssisted-by: llama.cpp:DeepSeek-v4-Flash-0731\n\n* ci : pass --repo to gh cache commands in ccache-clear\n\nIn container jobs gh cannot auto-detect the repository from git, so\ngh cache list/delete fail with 'failed to run git: not a git repository'.\nPass the repository explicitly via --repo using GITHUB_REPOSITORY.\n\nAssisted-by: llama.cpp:DeepSeek-v4-Flash-0731\n\n* ci : drop -new suffix from vulkan ccache key\n\nThe -new suffix was only needed to force a fresh cache. With\nccache-clear now evicting stale caches, the original key can be used\nagain. The old ccache-vulkan-ubuntu-24.04-arm-new entries still match\nthe ccache-clear key prefix and are cleaned up automatically.\n\nAssisted-by: llama.cpp:DeepSeek-v4-Flash-0731\n\n* ci : fix ccache-clear date parsing on macOS (BSD date)\n\nmacOS ships BSD date, which has no -d option. The older cutoff check\nwas silently disabled there: 'date: illegal option -- d' errors in the\nlog and the loop was only stopped by the min limit, risking deletion\nof caches not older than the cutoff (e.g. saved by a concurrent job).\n\nParse the ISO-8601 timestamps with GNU date when available and fall\nback to BSD date otherwise (TZ=UTC, fractional seconds dropped).\n\nAssisted-by: llama.cpp:DeepSeek-v4-Flash-0731\n\n* ci : extract ccache-clear logic into scripts/ccache-clear.sh\n\nThe composite action now consists of a dedicated step that installs the\nGitHub CLI when missing (e.g. in container jobs) and a thin step that\ncalls the new script. The script follows the make-release-checks.sh\nconventions (usage/env header, set -euo pipefail, CLI flags) and only\nchecks that gh is available. The action inputs are unchanged, so the\nworkflow steps are untouched.\n\nAssisted-by: llama.cpp:DeepSeek-v4-Flash-0731\n\n* ci : remove unused apple ccaches","tree":"0f6df18eb929856daca94b05ada439528798dd6a"}},{"name":"build-riscv.yml","mode":"0100644","size":6047,"lastCommit":{"hash":"491c4d7d2e2378c81fa0eb2bf916b437bbee4b0f","oid":"491c4d7d2e2378c81fa0eb2bf916b437bbee4b0f","parents":["939a7dd64837dc4228326c6a29d76a7552a707ed"],"author":{"name":"Georgi Gerganov","email":"ggerganov@gmail.com","when":"2026-05-28T09:44:25+03:00"},"committer":{"name":"GitHub","email":"noreply@github.com","when":"2026-05-28T09:44:25+03:00"},"message":"ci : refactor (#23789)\n\n* ci : separate CUDA windows workflow + fix names\n\n* ci : rename workflow\n\n* ci : prefix cache names with workflow name\n\n* ci : rename build.yml -> build-cpu.yml\n\n* ci : cache keys\n\n* ci : fix windows cuda/hip concurrency of release workflow\n\n* ci : fix apple cache names\n\n* ci : add TODOs\n\n* cont : keep just the last cache\n\n* ci : update release concurrency to queue\n\n* ci : move the release trigger to ubuntu-slim\n\n* ci : hip add TODO\n\n* cont : improve words\n\nCo-authored-by: Georgi Gerganov ","tree":"6732976e54ca3951492e4c6da812b6f96926ab26"}},{"name":"build-sanitize.yml","mode":"0100644","size":3051,"lastCommit":{"hash":"84e908c625fb60992b4cdef8180fb12fa9b4c4bf","oid":"84e908c625fb60992b4cdef8180fb12fa9b4c4bf","parents":["9558fa44c92746a58dd07ad1bf0c889715b938a6"],"author":{"name":"Eve","email":"139727413+netrunnereve@users.noreply.github.com","when":"2026-08-12T16:01:12Z"},"committer":{"name":"GitHub","email":"noreply@github.com","when":"2026-08-12T19:01:12+03:00"},"message":"ci: fix thread sanitizer + remove ccache (#26927)\n\n* test address on Intel-LNL-U7-258V\n\n* retry\n\n* run address on github\n\n* use native build for cpu\n\n* this should be runnable everywhere multicore\n\n* disable ccache\n\n---------\n\nCo-authored-by: Sigbjørn Skjæret ","tree":"cfc216317163e608580682b0167171779a673f4c"}},{"name":"build-self-hosted.yml","mode":"0100644","size":11779,"lastCommit":{"hash":"9d77fa17254e1dee4b9e92504c91611a60b1359f","oid":"9d77fa17254e1dee4b9e92504c91611a60b1359f","parents":["c0296022f361beaab4c6ecf5def391d2fc598dae"],"author":{"name":"Zijun Yu","email":"zijun.yu@intel.com","when":"2026-08-18T18:02:22+08:00"},"committer":{"name":"GitHub","email":"noreply@github.com","when":"2026-08-18T12:02:22+02:00"},"message":"ci : Update OpenVINO to 2026.3, skip nemotron-h rollback test (#27292)\n\n* update to ov-2026.3, update device drivers\n\n* ci: skip nemotron-h rollback test on OpenVINO\n\nThe OpenVINO backend does not support SSM_SCAN, so the Nemotron-H recurrent state rollback graph is split and cannot preserve the recurrent cache output shape. Keep the test enabled for other backends and retain the qwen35 OpenVINO rollback coverage.\n\n---------\n\nCo-authored-by: ravi9 ","tree":"4209b7a0a8180a2c50a1c605b8b913e52798a9cd"}},{"name":"build-sycl.yml","mode":"0100644","size":5209,"lastCommit":{"hash":"985b14912bc9a3d8f27c1ac569e46f4092123722","oid":"985b14912bc9a3d8f27c1ac569e46f4092123722","parents":["6036c635e2d1e9aeca11d01a84188c9db9ea1e2a"],"author":{"name":"Georgi Gerganov","email":"ggerganov@gmail.com","when":"2026-08-24T10:49:20+03:00"},"committer":{"name":"GitHub","email":"noreply@github.com","when":"2026-08-24T10:49:20+03:00"},"message":"ci : apply ccache-clear with older/min/dry-run to all ccache jobs (#27602)\n\n* ci : apply ccache-clear with older/min/dry-run to all ccache jobs\n\nAssisted-by: llama.cpp:DeepSeek-v4-Flash-0731\n\n* ci : install gh in ccache-clear if missing (container jobs)\n\nThe ccache-clear action relies on the gh CLI, which is not present in\ncontainer-based jobs. Install it on demand so those jobs can clear caches.\n\nAssisted-by: llama.cpp:DeepSeek-v4-Flash-0731\n\n* ci : install gh via apt repo in ccache-clear\n\nThe install.sh script used previously is no longer served (404). Switch to\nthe official GitHub CLI apt repository, which is still available.\n\nAssisted-by: llama.cpp:DeepSeek-v4-Flash-0731\n\n* ci : pass --repo to gh cache commands in ccache-clear\n\nIn container jobs gh cannot auto-detect the repository from git, so\ngh cache list/delete fail with 'failed to run git: not a git repository'.\nPass the repository explicitly via --repo using GITHUB_REPOSITORY.\n\nAssisted-by: llama.cpp:DeepSeek-v4-Flash-0731\n\n* ci : drop -new suffix from vulkan ccache key\n\nThe -new suffix was only needed to force a fresh cache. With\nccache-clear now evicting stale caches, the original key can be used\nagain. The old ccache-vulkan-ubuntu-24.04-arm-new entries still match\nthe ccache-clear key prefix and are cleaned up automatically.\n\nAssisted-by: llama.cpp:DeepSeek-v4-Flash-0731\n\n* ci : fix ccache-clear date parsing on macOS (BSD date)\n\nmacOS ships BSD date, which has no -d option. The older cutoff check\nwas silently disabled there: 'date: illegal option -- d' errors in the\nlog and the loop was only stopped by the min limit, risking deletion\nof caches not older than the cutoff (e.g. saved by a concurrent job).\n\nParse the ISO-8601 timestamps with GNU date when available and fall\nback to BSD date otherwise (TZ=UTC, fractional seconds dropped).\n\nAssisted-by: llama.cpp:DeepSeek-v4-Flash-0731\n\n* ci : extract ccache-clear logic into scripts/ccache-clear.sh\n\nThe composite action now consists of a dedicated step that installs the\nGitHub CLI when missing (e.g. in container jobs) and a thin step that\ncalls the new script. The script follows the make-release-checks.sh\nconventions (usage/env header, set -euo pipefail, CLI flags) and only\nchecks that gh is available. The action inputs are unchanged, so the\nworkflow steps are untouched.\n\nAssisted-by: llama.cpp:DeepSeek-v4-Flash-0731\n\n* ci : remove unused apple ccaches","tree":"0f6df18eb929856daca94b05ada439528798dd6a"}},{"name":"build-virtgpu.yml","mode":"0100644","size":1134,"lastCommit":{"hash":"928b486b0c8ef4a126086e078126cdb42e977fc7","oid":"928b486b0c8ef4a126086e078126cdb42e977fc7","parents":["7dbb0e998a125973091914aec2928a5104b36725"],"author":{"name":"Kevin Pouget","email":"kpouget@redhat.com","when":"2026-05-11T15:38:22+02:00"},"committer":{"name":"GitHub","email":"noreply@github.com","when":"2026-05-11T21:38:22+08:00"},"message":"ggml-virtgpu: Add a GHA build check (#22943)\n\n* [ggml-virtgpu] Add a GHA build check\n\n* Apply suggestions from code review\n\nCo-authored-by: Sigbjørn Skjæret \n\n---------\n\nCo-authored-by: Sigbjørn Skjæret ","tree":"aaf64af1660496d030efd46317e36e174ddb15a5"}},{"name":"build-vulkan.yml","mode":"0100644","size":6191,"lastCommit":{"hash":"985b14912bc9a3d8f27c1ac569e46f4092123722","oid":"985b14912bc9a3d8f27c1ac569e46f4092123722","parents":["6036c635e2d1e9aeca11d01a84188c9db9ea1e2a"],"author":{"name":"Georgi Gerganov","email":"ggerganov@gmail.com","when":"2026-08-24T10:49:20+03:00"},"committer":{"name":"GitHub","email":"noreply@github.com","when":"2026-08-24T10:49:20+03:00"},"message":"ci : apply ccache-clear with older/min/dry-run to all ccache jobs (#27602)\n\n* ci : apply ccache-clear with older/min/dry-run to all ccache jobs\n\nAssisted-by: llama.cpp:DeepSeek-v4-Flash-0731\n\n* ci : install gh in ccache-clear if missing (container jobs)\n\nThe ccache-clear action relies on the gh CLI, which is not present in\ncontainer-based jobs. Install it on demand so those jobs can clear caches.\n\nAssisted-by: llama.cpp:DeepSeek-v4-Flash-0731\n\n* ci : install gh via apt repo in ccache-clear\n\nThe install.sh script used previously is no longer served (404). Switch to\nthe official GitHub CLI apt repository, which is still available.\n\nAssisted-by: llama.cpp:DeepSeek-v4-Flash-0731\n\n* ci : pass --repo to gh cache commands in ccache-clear\n\nIn container jobs gh cannot auto-detect the repository from git, so\ngh cache list/delete fail with 'failed to run git: not a git repository'.\nPass the repository explicitly via --repo using GITHUB_REPOSITORY.\n\nAssisted-by: llama.cpp:DeepSeek-v4-Flash-0731\n\n* ci : drop -new suffix from vulkan ccache key\n\nThe -new suffix was only needed to force a fresh cache. With\nccache-clear now evicting stale caches, the original key can be used\nagain. The old ccache-vulkan-ubuntu-24.04-arm-new entries still match\nthe ccache-clear key prefix and are cleaned up automatically.\n\nAssisted-by: llama.cpp:DeepSeek-v4-Flash-0731\n\n* ci : fix ccache-clear date parsing on macOS (BSD date)\n\nmacOS ships BSD date, which has no -d option. The older cutoff check\nwas silently disabled there: 'date: illegal option -- d' errors in the\nlog and the loop was only stopped by the min limit, risking deletion\nof caches not older than the cutoff (e.g. saved by a concurrent job).\n\nParse the ISO-8601 timestamps with GNU date when available and fall\nback to BSD date otherwise (TZ=UTC, fractional seconds dropped).\n\nAssisted-by: llama.cpp:DeepSeek-v4-Flash-0731\n\n* ci : extract ccache-clear logic into scripts/ccache-clear.sh\n\nThe composite action now consists of a dedicated step that installs the\nGitHub CLI when missing (e.g. in container jobs) and a thin step that\ncalls the new script. The script follows the make-release-checks.sh\nconventions (usage/env header, set -euo pipefail, CLI flags) and only\nchecks that gh is available. The action inputs are unchanged, so the\nworkflow steps are untouched.\n\nAssisted-by: llama.cpp:DeepSeek-v4-Flash-0731\n\n* ci : remove unused apple ccaches","tree":"0f6df18eb929856daca94b05ada439528798dd6a"}},{"name":"build-wasm.yml","mode":"0100644","size":2618,"lastCommit":{"hash":"985b14912bc9a3d8f27c1ac569e46f4092123722","oid":"985b14912bc9a3d8f27c1ac569e46f4092123722","parents":["6036c635e2d1e9aeca11d01a84188c9db9ea1e2a"],"author":{"name":"Georgi Gerganov","email":"ggerganov@gmail.com","when":"2026-08-24T10:49:20+03:00"},"committer":{"name":"GitHub","email":"noreply@github.com","when":"2026-08-24T10:49:20+03:00"},"message":"ci : apply ccache-clear with older/min/dry-run to all ccache jobs (#27602)\n\n* ci : apply ccache-clear with older/min/dry-run to all ccache jobs\n\nAssisted-by: llama.cpp:DeepSeek-v4-Flash-0731\n\n* ci : install gh in ccache-clear if missing (container jobs)\n\nThe ccache-clear action relies on the gh CLI, which is not present in\ncontainer-based jobs. Install it on demand so those jobs can clear caches.\n\nAssisted-by: llama.cpp:DeepSeek-v4-Flash-0731\n\n* ci : install gh via apt repo in ccache-clear\n\nThe install.sh script used previously is no longer served (404). Switch to\nthe official GitHub CLI apt repository, which is still available.\n\nAssisted-by: llama.cpp:DeepSeek-v4-Flash-0731\n\n* ci : pass --repo to gh cache commands in ccache-clear\n\nIn container jobs gh cannot auto-detect the repository from git, so\ngh cache list/delete fail with 'failed to run git: not a git repository'.\nPass the repository explicitly via --repo using GITHUB_REPOSITORY.\n\nAssisted-by: llama.cpp:DeepSeek-v4-Flash-0731\n\n* ci : drop -new suffix from vulkan ccache key\n\nThe -new suffix was only needed to force a fresh cache. With\nccache-clear now evicting stale caches, the original key can be used\nagain. The old ccache-vulkan-ubuntu-24.04-arm-new entries still match\nthe ccache-clear key prefix and are cleaned up automatically.\n\nAssisted-by: llama.cpp:DeepSeek-v4-Flash-0731\n\n* ci : fix ccache-clear date parsing on macOS (BSD date)\n\nmacOS ships BSD date, which has no -d option. The older cutoff check\nwas silently disabled there: 'date: illegal option -- d' errors in the\nlog and the loop was only stopped by the min limit, risking deletion\nof caches not older than the cutoff (e.g. saved by a concurrent job).\n\nParse the ISO-8601 timestamps with GNU date when available and fall\nback to BSD date otherwise (TZ=UTC, fractional seconds dropped).\n\nAssisted-by: llama.cpp:DeepSeek-v4-Flash-0731\n\n* ci : extract ccache-clear logic into scripts/ccache-clear.sh\n\nThe composite action now consists of a dedicated step that installs the\nGitHub CLI when missing (e.g. in container jobs) and a thin step that\ncalls the new script. The script follows the make-release-checks.sh\nconventions (usage/env header, set -euo pipefail, CLI flags) and only\nchecks that gh is available. The action inputs are unchanged, so the\nworkflow steps are untouched.\n\nAssisted-by: llama.cpp:DeepSeek-v4-Flash-0731\n\n* ci : remove unused apple ccaches","tree":"0f6df18eb929856daca94b05ada439528798dd6a"}},{"name":"build-webgpu.yml","mode":"0100644","size":5443,"lastCommit":{"hash":"985b14912bc9a3d8f27c1ac569e46f4092123722","oid":"985b14912bc9a3d8f27c1ac569e46f4092123722","parents":["6036c635e2d1e9aeca11d01a84188c9db9ea1e2a"],"author":{"name":"Georgi Gerganov","email":"ggerganov@gmail.com","when":"2026-08-24T10:49:20+03:00"},"committer":{"name":"GitHub","email":"noreply@github.com","when":"2026-08-24T10:49:20+03:00"},"message":"ci : apply ccache-clear with older/min/dry-run to all ccache jobs (#27602)\n\n* ci : apply ccache-clear with older/min/dry-run to all ccache jobs\n\nAssisted-by: llama.cpp:DeepSeek-v4-Flash-0731\n\n* ci : install gh in ccache-clear if missing (container jobs)\n\nThe ccache-clear action relies on the gh CLI, which is not present in\ncontainer-based jobs. Install it on demand so those jobs can clear caches.\n\nAssisted-by: llama.cpp:DeepSeek-v4-Flash-0731\n\n* ci : install gh via apt repo in ccache-clear\n\nThe install.sh script used previously is no longer served (404). Switch to\nthe official GitHub CLI apt repository, which is still available.\n\nAssisted-by: llama.cpp:DeepSeek-v4-Flash-0731\n\n* ci : pass --repo to gh cache commands in ccache-clear\n\nIn container jobs gh cannot auto-detect the repository from git, so\ngh cache list/delete fail with 'failed to run git: not a git repository'.\nPass the repository explicitly via --repo using GITHUB_REPOSITORY.\n\nAssisted-by: llama.cpp:DeepSeek-v4-Flash-0731\n\n* ci : drop -new suffix from vulkan ccache key\n\nThe -new suffix was only needed to force a fresh cache. With\nccache-clear now evicting stale caches, the original key can be used\nagain. The old ccache-vulkan-ubuntu-24.04-arm-new entries still match\nthe ccache-clear key prefix and are cleaned up automatically.\n\nAssisted-by: llama.cpp:DeepSeek-v4-Flash-0731\n\n* ci : fix ccache-clear date parsing on macOS (BSD date)\n\nmacOS ships BSD date, which has no -d option. The older cutoff check\nwas silently disabled there: 'date: illegal option -- d' errors in the\nlog and the loop was only stopped by the min limit, risking deletion\nof caches not older than the cutoff (e.g. saved by a concurrent job).\n\nParse the ISO-8601 timestamps with GNU date when available and fall\nback to BSD date otherwise (TZ=UTC, fractional seconds dropped).\n\nAssisted-by: llama.cpp:DeepSeek-v4-Flash-0731\n\n* ci : extract ccache-clear logic into scripts/ccache-clear.sh\n\nThe composite action now consists of a dedicated step that installs the\nGitHub CLI when missing (e.g. in container jobs) and a thin step that\ncalls the new script. The script follows the make-release-checks.sh\nconventions (usage/env header, set -euo pipefail, CLI flags) and only\nchecks that gh is available. The action inputs are unchanged, so the\nworkflow steps are untouched.\n\nAssisted-by: llama.cpp:DeepSeek-v4-Flash-0731\n\n* ci : remove unused apple ccaches","tree":"0f6df18eb929856daca94b05ada439528798dd6a"}},{"name":"check-vendor.yml","mode":"0100644","size":1232,"lastCommit":{"hash":"28123a3937f7d056a4dded43752b216507384373","oid":"28123a3937f7d056a4dded43752b216507384373","parents":["549b9d84330c327e6791fa812a7d60c0cf63572e"],"author":{"name":"Georgi Gerganov","email":"ggerganov@gmail.com","when":"2026-05-25T08:11:19+03:00"},"committer":{"name":"GitHub","email":"noreply@github.com","when":"2026-05-25T08:11:19+03:00"},"message":"ci : move most slim jobs to self-hosted runners (#23619)\n\n* ci : remove tag from build-self-hosted.yml\n\n* ci : slim -> self-hosted\n\n* ci : prevent heavy CPU jobs from running on fast runners\n\n* ci : prevent cmake pkg to run on dedicated fast runners\n\n* ci : try to bump 3.11 -> 3.13\n\n* ci : move lint back to 3.11\n\n* ci : back to 3.11\n\n* ci : add comment about UI jobs\n\n* ci : move python requirements check to CPU runners\n\nthis job is a bit slow for a dedicated \"fast\" runner\n\n* ci : add self-hosted ui workflow\n\n* ci : fix UI naming\n\n* tmp to check if arm64 fast is compatible with all jobs\n\n* revert last commit","tree":"1dd08661bb865152283614cc888549d7516459c4"}},{"name":"close-issue.yml","mode":"0100644","size":954,"lastCommit":{"hash":"a8bad3842e92f61dadcd71417f6e10b68bff0ee4","oid":"a8bad3842e92f61dadcd71417f6e10b68bff0ee4","parents":["75f3bc94e649616162981c322e8e6b88ca5491e8"],"author":{"name":"Christian Kastner","email":"ckk@kvr.at","when":"2026-04-13T19:18:44+02:00"},"committer":{"name":"GitHub","email":"noreply@github.com","when":"2026-04-14T01:18:44+08:00"},"message":"ci: Also exempt 'security' tag from auto-close (#21844)","tree":"dd6fdbffcf8c9f0d6a40e337ce3c159167fef725"}},{"name":"code-style.yml","mode":"0100644","size":1741,"lastCommit":{"hash":"28123a3937f7d056a4dded43752b216507384373","oid":"28123a3937f7d056a4dded43752b216507384373","parents":["549b9d84330c327e6791fa812a7d60c0cf63572e"],"author":{"name":"Georgi Gerganov","email":"ggerganov@gmail.com","when":"2026-05-25T08:11:19+03:00"},"committer":{"name":"GitHub","email":"noreply@github.com","when":"2026-05-25T08:11:19+03:00"},"message":"ci : move most slim jobs to self-hosted runners (#23619)\n\n* ci : remove tag from build-self-hosted.yml\n\n* ci : slim -> self-hosted\n\n* ci : prevent heavy CPU jobs from running on fast runners\n\n* ci : prevent cmake pkg to run on dedicated fast runners\n\n* ci : try to bump 3.11 -> 3.13\n\n* ci : move lint back to 3.11\n\n* ci : back to 3.11\n\n* ci : add comment about UI jobs\n\n* ci : move python requirements check to CPU runners\n\nthis job is a bit slow for a dedicated \"fast\" runner\n\n* ci : add self-hosted ui workflow\n\n* ci : fix UI naming\n\n* tmp to check if arm64 fast is compatible with all jobs\n\n* revert last commit","tree":"1dd08661bb865152283614cc888549d7516459c4"}},{"name":"copilot-setup-steps.yml","mode":"0100644","size":2180,"lastCommit":{"hash":"8fc85db9d2ea0951191ac31d7777781a23017037","oid":"8fc85db9d2ea0951191ac31d7777781a23017037","parents":["3a60d06ad9477dae581bb911bd699350802c3cea"],"author":{"name":"Sigbjørn Skjæret","email":"sigbjorn.skjaeret@scala.com","when":"2026-03-25T09:55:37+01:00"},"committer":{"name":"GitHub","email":"noreply@github.com","when":"2026-03-25T10:55:37+02:00"},"message":"ci : limit requirements versions (#20980)\n\n* set requests version\n\n* limit versions outside requirements","tree":"b7c63136e078485e2f79eb49286933d7adf81864"}},{"name":"docker.yml","mode":"0100644","size":25925,"lastCommit":{"hash":"0adcc3bb571011bff8b91335d0728a82845c421b","oid":"0adcc3bb571011bff8b91335d0728a82845c421b","parents":["79fe79919436fa26029a0ac1ea25eb34c2c2fb0d"],"author":{"name":"Jeremie Miller","email":"jeremie.miller@gmail.com","when":"2026-08-19T01:23:52-06:00"},"committer":{"name":"GitHub","email":"noreply@github.com","when":"2026-08-19T10:23:52+03:00"},"message":"ci : add attestation for signed release artifacts (#25933)","tree":"87f6dbbe2eb76cdda777527c2ee83b4ab5efb55f"}},{"name":"editorconfig.yml","mode":"0100644","size":577,"lastCommit":{"hash":"28123a3937f7d056a4dded43752b216507384373","oid":"28123a3937f7d056a4dded43752b216507384373","parents":["549b9d84330c327e6791fa812a7d60c0cf63572e"],"author":{"name":"Georgi Gerganov","email":"ggerganov@gmail.com","when":"2026-05-25T08:11:19+03:00"},"committer":{"name":"GitHub","email":"noreply@github.com","when":"2026-05-25T08:11:19+03:00"},"message":"ci : move most slim jobs to self-hosted runners (#23619)\n\n* ci : remove tag from build-self-hosted.yml\n\n* ci : slim -> self-hosted\n\n* ci : prevent heavy CPU jobs from running on fast runners\n\n* ci : prevent cmake pkg to run on dedicated fast runners\n\n* ci : try to bump 3.11 -> 3.13\n\n* ci : move lint back to 3.11\n\n* ci : back to 3.11\n\n* ci : add comment about UI jobs\n\n* ci : move python requirements check to CPU runners\n\nthis job is a bit slow for a dedicated \"fast\" runner\n\n* ci : add self-hosted ui workflow\n\n* ci : fix UI naming\n\n* tmp to check if arm64 fast is compatible with all jobs\n\n* revert last commit","tree":"1dd08661bb865152283614cc888549d7516459c4"}},{"name":"gguf-publish.yml","mode":"0100644","size":1272,"lastCommit":{"hash":"a290ce626663dae1d54f70bce3ca6d8f67aab62f","oid":"a290ce626663dae1d54f70bce3ca6d8f67aab62f","parents":["a00e47e422bc4e48b8d2cdcfb16b7e55748237c2"],"author":{"name":"Georgi Gerganov","email":"ggerganov@gmail.com","when":"2026-05-06T15:46:14+03:00"},"committer":{"name":"GitHub","email":"noreply@github.com","when":"2026-05-06T14:46:14+02:00"},"message":"gguf-py : bump version to 0.19.0 (#22664)\n\n* gguf-py : bump version to 0.19.0\n\n* bump poetry\n\n---------\n\nCo-authored-by: Sigbjørn Skjæret ","tree":"37b937419b59dcda1749672655598a9344640a42"}},{"name":"hip-quality-check.yml","mode":"0100644","size":2807,"lastCommit":{"hash":"985b14912bc9a3d8f27c1ac569e46f4092123722","oid":"985b14912bc9a3d8f27c1ac569e46f4092123722","parents":["6036c635e2d1e9aeca11d01a84188c9db9ea1e2a"],"author":{"name":"Georgi Gerganov","email":"ggerganov@gmail.com","when":"2026-08-24T10:49:20+03:00"},"committer":{"name":"GitHub","email":"noreply@github.com","when":"2026-08-24T10:49:20+03:00"},"message":"ci : apply ccache-clear with older/min/dry-run to all ccache jobs (#27602)\n\n* ci : apply ccache-clear with older/min/dry-run to all ccache jobs\n\nAssisted-by: llama.cpp:DeepSeek-v4-Flash-0731\n\n* ci : install gh in ccache-clear if missing (container jobs)\n\nThe ccache-clear action relies on the gh CLI, which is not present in\ncontainer-based jobs. Install it on demand so those jobs can clear caches.\n\nAssisted-by: llama.cpp:DeepSeek-v4-Flash-0731\n\n* ci : install gh via apt repo in ccache-clear\n\nThe install.sh script used previously is no longer served (404). Switch to\nthe official GitHub CLI apt repository, which is still available.\n\nAssisted-by: llama.cpp:DeepSeek-v4-Flash-0731\n\n* ci : pass --repo to gh cache commands in ccache-clear\n\nIn container jobs gh cannot auto-detect the repository from git, so\ngh cache list/delete fail with 'failed to run git: not a git repository'.\nPass the repository explicitly via --repo using GITHUB_REPOSITORY.\n\nAssisted-by: llama.cpp:DeepSeek-v4-Flash-0731\n\n* ci : drop -new suffix from vulkan ccache key\n\nThe -new suffix was only needed to force a fresh cache. With\nccache-clear now evicting stale caches, the original key can be used\nagain. The old ccache-vulkan-ubuntu-24.04-arm-new entries still match\nthe ccache-clear key prefix and are cleaned up automatically.\n\nAssisted-by: llama.cpp:DeepSeek-v4-Flash-0731\n\n* ci : fix ccache-clear date parsing on macOS (BSD date)\n\nmacOS ships BSD date, which has no -d option. The older cutoff check\nwas silently disabled there: 'date: illegal option -- d' errors in the\nlog and the loop was only stopped by the min limit, risking deletion\nof caches not older than the cutoff (e.g. saved by a concurrent job).\n\nParse the ISO-8601 timestamps with GNU date when available and fall\nback to BSD date otherwise (TZ=UTC, fractional seconds dropped).\n\nAssisted-by: llama.cpp:DeepSeek-v4-Flash-0731\n\n* ci : extract ccache-clear logic into scripts/ccache-clear.sh\n\nThe composite action now consists of a dedicated step that installs the\nGitHub CLI when missing (e.g. in container jobs) and a thin step that\ncalls the new script. The script follows the make-release-checks.sh\nconventions (usage/env header, set -euo pipefail, CLI flags) and only\nchecks that gh is available. The action inputs are unchanged, so the\nworkflow steps are untouched.\n\nAssisted-by: llama.cpp:DeepSeek-v4-Flash-0731\n\n* ci : remove unused apple ccaches","tree":"0f6df18eb929856daca94b05ada439528798dd6a"}},{"name":"labeler.yml","mode":"0100644","size":352,"lastCommit":{"hash":"142cbe2ac68978e5dec3a2e19c1b64ef1c5740b1","oid":"142cbe2ac68978e5dec3a2e19c1b64ef1c5740b1","parents":["56f3ebf38ea5757447e7ea167cda15e1eebf9601"],"author":{"name":"Sigbjørn Skjæret","email":"sigbjorn.skjaeret@scala.com","when":"2026-01-26T15:22:49+01:00"},"committer":{"name":"GitHub","email":"noreply@github.com","when":"2026-01-26T15:22:49+01:00"},"message":"ci : use new 1vCPU runner for lightweight jobs (#19107)\n\n* use new 1vCPU runner for lightweight jobs\n\n* pyright is too heavy, look into ty some day\n\nuse new pip-install input","tree":"33ac2dad5b1950aad92baeee3c3b5dc626047816"}},{"name":"make-release.yml","mode":"0100644","size":4454,"lastCommit":{"hash":"a3b9c23ead5054832c5ed4b43f652fa1123869a3","oid":"a3b9c23ead5054832c5ed4b43f652fa1123869a3","parents":["5a32f7b66ef6cfb3e60deea26e3454cc6ad3438c"],"author":{"name":"Georgi Gerganov","email":"ggerganov@gmail.com","when":"2026-08-21T21:41:25+03:00"},"committer":{"name":"GitHub","email":"noreply@github.com","when":"2026-08-21T21:41:25+03:00"},"message":"ci : fix empty release_id in make-release upload step (#27516)\n\nThe 'Create release' step had no id, so steps.create_release.outputs.id\nresolved to an empty string in the 'Upload nightly-tag.txt' step. The\nuploadReleaseAsset call then hit /releases//assets and failed with HTTP\n404 (Unhandled error: HttpError), e.g. run 32513839499.\n\nAdd id: create_release to the step; the action already exposes the id\noutput.\n\nAssisted-by: pi:llama.cpp/Qwen3.8-27B","tree":"427f477502136c56f7f1c4d7153834415adbecbf"}},{"name":"pr-draft-label.yml","mode":"0100644","size":638,"lastCommit":{"hash":"936918514ce522b553c0fd80b169a6440e6096c6","oid":"936918514ce522b553c0fd80b169a6440e6096c6","parents":["08659901c43b51de735740f1cf61bb82fbe0c4e4"],"author":{"name":"Xuan-Son Nguyen","email":"son@huggingface.co","when":"2026-08-09T16:51:21+02:00"},"committer":{"name":"GitHub","email":"noreply@github.com","when":"2026-08-09T16:51:21+02:00"},"message":"ci: add pr-draft-label (#26801)","tree":"d6a63761406191af5bbb392e2a1d3e520454cfae"}},{"name":"pre-tokenizer-hashes.yml","mode":"0100644","size":1565,"lastCommit":{"hash":"062d3115aa12267d34d7ac5b78bd5d0ad48778ad","oid":"062d3115aa12267d34d7ac5b78bd5d0ad48778ad","parents":["314e729347defd9851e857f78084160c5786a7d8"],"author":{"name":"Sigbjørn Skjæret","email":"sigbjorn.skjaeret@scala.com","when":"2026-05-25T10:41:25+02:00"},"committer":{"name":"GitHub","email":"noreply@github.com","when":"2026-05-25T10:41:25+02:00"},"message":"ci : fix pre-tokenizer-hashes check (#23651)","tree":"93a23544df882f2bb6b37b0197d7ae319f812a0c"}},{"name":"python-check-requirements.yml","mode":"0100644","size":927,"lastCommit":{"hash":"28123a3937f7d056a4dded43752b216507384373","oid":"28123a3937f7d056a4dded43752b216507384373","parents":["549b9d84330c327e6791fa812a7d60c0cf63572e"],"author":{"name":"Georgi Gerganov","email":"ggerganov@gmail.com","when":"2026-05-25T08:11:19+03:00"},"committer":{"name":"GitHub","email":"noreply@github.com","when":"2026-05-25T08:11:19+03:00"},"message":"ci : move most slim jobs to self-hosted runners (#23619)\n\n* ci : remove tag from build-self-hosted.yml\n\n* ci : slim -> self-hosted\n\n* ci : prevent heavy CPU jobs from running on fast runners\n\n* ci : prevent cmake pkg to run on dedicated fast runners\n\n* ci : try to bump 3.11 -> 3.13\n\n* ci : move lint back to 3.11\n\n* ci : back to 3.11\n\n* ci : add comment about UI jobs\n\n* ci : move python requirements check to CPU runners\n\nthis job is a bit slow for a dedicated \"fast\" runner\n\n* ci : add self-hosted ui workflow\n\n* ci : fix UI naming\n\n* tmp to check if arm64 fast is compatible with all jobs\n\n* revert last commit","tree":"1dd08661bb865152283614cc888549d7516459c4"}},{"name":"python-lint.yml","mode":"0100644","size":841,"lastCommit":{"hash":"28123a3937f7d056a4dded43752b216507384373","oid":"28123a3937f7d056a4dded43752b216507384373","parents":["549b9d84330c327e6791fa812a7d60c0cf63572e"],"author":{"name":"Georgi Gerganov","email":"ggerganov@gmail.com","when":"2026-05-25T08:11:19+03:00"},"committer":{"name":"GitHub","email":"noreply@github.com","when":"2026-05-25T08:11:19+03:00"},"message":"ci : move most slim jobs to self-hosted runners (#23619)\n\n* ci : remove tag from build-self-hosted.yml\n\n* ci : slim -> self-hosted\n\n* ci : prevent heavy CPU jobs from running on fast runners\n\n* ci : prevent cmake pkg to run on dedicated fast runners\n\n* ci : try to bump 3.11 -> 3.13\n\n* ci : move lint back to 3.11\n\n* ci : back to 3.11\n\n* ci : add comment about UI jobs\n\n* ci : move python requirements check to CPU runners\n\nthis job is a bit slow for a dedicated \"fast\" runner\n\n* ci : add self-hosted ui workflow\n\n* ci : fix UI naming\n\n* tmp to check if arm64 fast is compatible with all jobs\n\n* revert last commit","tree":"1dd08661bb865152283614cc888549d7516459c4"}},{"name":"python-type-check.yml","mode":"0100644","size":1142,"lastCommit":{"hash":"28123a3937f7d056a4dded43752b216507384373","oid":"28123a3937f7d056a4dded43752b216507384373","parents":["549b9d84330c327e6791fa812a7d60c0cf63572e"],"author":{"name":"Georgi Gerganov","email":"ggerganov@gmail.com","when":"2026-05-25T08:11:19+03:00"},"committer":{"name":"GitHub","email":"noreply@github.com","when":"2026-05-25T08:11:19+03:00"},"message":"ci : move most slim jobs to self-hosted runners (#23619)\n\n* ci : remove tag from build-self-hosted.yml\n\n* ci : slim -> self-hosted\n\n* ci : prevent heavy CPU jobs from running on fast runners\n\n* ci : prevent cmake pkg to run on dedicated fast runners\n\n* ci : try to bump 3.11 -> 3.13\n\n* ci : move lint back to 3.11\n\n* ci : back to 3.11\n\n* ci : add comment about UI jobs\n\n* ci : move python requirements check to CPU runners\n\nthis job is a bit slow for a dedicated \"fast\" runner\n\n* ci : add self-hosted ui workflow\n\n* ci : fix UI naming\n\n* tmp to check if arm64 fast is compatible with all jobs\n\n* revert last commit","tree":"1dd08661bb865152283614cc888549d7516459c4"}},{"name":"release-prism.yml","mode":"0100644","size":37143,"lastCommit":{"hash":"9a4dd89d0a6b609e21cd6f43cc4c057f05b356cc","oid":"9a4dd89d0a6b609e21cd6f43cc4c057f05b356cc","parents":["842b1880415d6f508f03b789e5ce70194def7bfd"],"author":{"name":"bri-prism","email":"288398250+bri-prism@users.noreply.github.com","when":"2026-09-25T10:39:48-07:00"},"committer":{"name":"GitHub","email":"noreply@github.com","when":"2026-09-25T10:39:48-07:00"},"message":"release: ship CPU backend variants in the CUDA and HIP bundles (#241) (#271)\n\nThe Windows CUDA x64, Linux CUDA and Windows HIP release jobs build with\nGGML_NATIVE=OFF but without GGML_BACKEND_DL/GGML_CPU_ALL_VARIANTS. With\nnative detection off and no explicit ISA flags, ggml enables AVX/AVX2/FMA/F16C\nby default, so each bundle ships a single AVX2-only ggml-cpu library. On a CPU\nwithout AVX2 the process dies at startup with no message: even\n`llama-server --version` exits silently (#241). The CPU and Vulkan bundles\nalready build every variant and pick one at runtime (Vulkan since #198).\n\nBuild those three bundles with GGML_BACKEND_DL=ON and GGML_CPU_ALL_VARIANTS=ON\nas well. The arm64 CUDA leg already used BACKEND_DL with GGML_CPU=OFF and is\nunchanged.\n\n\nClaude-Session: https://claude.ai/code/session_012Q6ecxozUXFFFLE3N2SMiR\n\nCo-authored-by: Claude Opus 5.5 (1M context) ","tree":"2a993eac47a68f5abf803968e229f6aeb150591f"}},{"name":"release.yml","mode":"0100644","size":67574,"lastCommit":{"hash":"e85caa81ea2b65797396018c179b87ad61fa38ab","oid":"e85caa81ea2b65797396018c179b87ad61fa38ab","parents":["2115b73d8ebdbd659075cce66c609506863bc826"],"author":{"name":"Mario Limonciello","email":"mario.limonciello@amd.com","when":"2026-08-22T05:28:30-05:00"},"committer":{"name":"GitHub","email":"noreply@github.com","when":"2026-08-22T13:28:30+03:00"},"message":"ci : Restore ROCm job for Ubuntu (#27399)\n\n* Revert \"ci : disable ubuntu-rocm (#26969)\"\n\nThis reverts commit 9558fa44c92746a58dd07ad1bf0c889715b938a6.\n\n* ci: set ccache compiler_check=content for ROCm build\n\nThe ROCm toolchain is pip-installed fresh on every run, so the clang binary's\nmtime changes each time. With ccache's default compiler_check=mtime that\ninvalidates the whole cache and warm builds only reached ~70% hits. Hash the\ncompiler contents instead so the cache survives toolchain reinstalls.\n\n* Update ccache size to 1GB\n\nWe're waivering with so many architectures built, we need a bigger\nccache limit.\n\n* merge fix\n\n---------\n\nCo-authored-by: Jim Wu \nCo-authored-by: Sigbjørn Skjæret ","tree":"85d6c1d06c249f1cdeecaf27ad1cb9fa3d48727f"}},{"name":"server-sanitize.yml","mode":"0100644","size":3505,"lastCommit":{"hash":"b3df57286c336255fd02c2162b0b5fe3adc2965e","oid":"b3df57286c336255fd02c2162b0b5fe3adc2965e","parents":["4801e3c567d5131dd41b387df5f2d4b1370d92be"],"author":{"name":"Georgi Gerganov","email":"ggerganov@gmail.com","when":"2026-08-11T09:07:13+03:00"},"committer":{"name":"GitHub","email":"noreply@github.com","when":"2026-08-11T09:07:13+03:00"},"message":"tests : clean-up server test, use `tests.sh` in ci (#26886)\n\n* tests : remove fetch_server_test_models.py\n\n* ci : use tests.sh wrapper of pytest","tree":"859ba17872de46803a72ef16edd55d76d793c1c1"}},{"name":"server-self-hosted.yml","mode":"0100644","size":6571,"lastCommit":{"hash":"b3df57286c336255fd02c2162b0b5fe3adc2965e","oid":"b3df57286c336255fd02c2162b0b5fe3adc2965e","parents":["4801e3c567d5131dd41b387df5f2d4b1370d92be"],"author":{"name":"Georgi Gerganov","email":"ggerganov@gmail.com","when":"2026-08-11T09:07:13+03:00"},"committer":{"name":"GitHub","email":"noreply@github.com","when":"2026-08-11T09:07:13+03:00"},"message":"tests : clean-up server test, use `tests.sh` in ci (#26886)\n\n* tests : remove fetch_server_test_models.py\n\n* ci : use tests.sh wrapper of pytest","tree":"859ba17872de46803a72ef16edd55d76d793c1c1"}},{"name":"server.yml","mode":"0100644","size":5544,"lastCommit":{"hash":"985b14912bc9a3d8f27c1ac569e46f4092123722","oid":"985b14912bc9a3d8f27c1ac569e46f4092123722","parents":["6036c635e2d1e9aeca11d01a84188c9db9ea1e2a"],"author":{"name":"Georgi Gerganov","email":"ggerganov@gmail.com","when":"2026-08-24T10:49:20+03:00"},"committer":{"name":"GitHub","email":"noreply@github.com","when":"2026-08-24T10:49:20+03:00"},"message":"ci : apply ccache-clear with older/min/dry-run to all ccache jobs (#27602)\n\n* ci : apply ccache-clear with older/min/dry-run to all ccache jobs\n\nAssisted-by: llama.cpp:DeepSeek-v4-Flash-0731\n\n* ci : install gh in ccache-clear if missing (container jobs)\n\nThe ccache-clear action relies on the gh CLI, which is not present in\ncontainer-based jobs. Install it on demand so those jobs can clear caches.\n\nAssisted-by: llama.cpp:DeepSeek-v4-Flash-0731\n\n* ci : install gh via apt repo in ccache-clear\n\nThe install.sh script used previously is no longer served (404). Switch to\nthe official GitHub CLI apt repository, which is still available.\n\nAssisted-by: llama.cpp:DeepSeek-v4-Flash-0731\n\n* ci : pass --repo to gh cache commands in ccache-clear\n\nIn container jobs gh cannot auto-detect the repository from git, so\ngh cache list/delete fail with 'failed to run git: not a git repository'.\nPass the repository explicitly via --repo using GITHUB_REPOSITORY.\n\nAssisted-by: llama.cpp:DeepSeek-v4-Flash-0731\n\n* ci : drop -new suffix from vulkan ccache key\n\nThe -new suffix was only needed to force a fresh cache. With\nccache-clear now evicting stale caches, the original key can be used\nagain. The old ccache-vulkan-ubuntu-24.04-arm-new entries still match\nthe ccache-clear key prefix and are cleaned up automatically.\n\nAssisted-by: llama.cpp:DeepSeek-v4-Flash-0731\n\n* ci : fix ccache-clear date parsing on macOS (BSD date)\n\nmacOS ships BSD date, which has no -d option. The older cutoff check\nwas silently disabled there: 'date: illegal option -- d' errors in the\nlog and the loop was only stopped by the min limit, risking deletion\nof caches not older than the cutoff (e.g. saved by a concurrent job).\n\nParse the ISO-8601 timestamps with GNU date when available and fall\nback to BSD date otherwise (TZ=UTC, fractional seconds dropped).\n\nAssisted-by: llama.cpp:DeepSeek-v4-Flash-0731\n\n* ci : extract ccache-clear logic into scripts/ccache-clear.sh\n\nThe composite action now consists of a dedicated step that installs the\nGitHub CLI when missing (e.g. in container jobs) and a thin step that\ncalls the new script. The script follows the make-release-checks.sh\nconventions (usage/env header, set -euo pipefail, CLI flags) and only\nchecks that gh is available. The action inputs are unchanged, so the\nworkflow steps are untouched.\n\nAssisted-by: llama.cpp:DeepSeek-v4-Flash-0731\n\n* ci : remove unused apple ccaches","tree":"0f6df18eb929856daca94b05ada439528798dd6a"}},{"name":"ui-build-self-hosted.yml","mode":"0100644","size":795,"lastCommit":{"hash":"57fe1f07c3b6a1de3f4fff19098e2056a85275b7","oid":"57fe1f07c3b6a1de3f4fff19098e2056a85275b7","parents":["d8a24ccee207a1ff24c513fe1c7d3222b3ccd837"],"author":{"name":"Xuan-Son Nguyen","email":"son@huggingface.co","when":"2026-06-13T11:51:20+02:00"},"committer":{"name":"GitHub","email":"noreply@github.com","when":"2026-06-13T11:51:20+02:00"},"message":"server: clean up static assets handling (#24550)\n\n* server: clean up static assets handling\n\n* nits\n\n* simplify file name handling, use static file name everywhere\n\n* cmake/ui : bundle UI assets in an archive\n\n* ui : run prettier on post-build.js\n\n---------\n\nCo-authored-by: Alde Rojas ","tree":"37d64d71e58068bbaf48b4f7cb964c9ba4db55c5"}},{"name":"ui-build.yml","mode":"0100644","size":1223,"lastCommit":{"hash":"57fe1f07c3b6a1de3f4fff19098e2056a85275b7","oid":"57fe1f07c3b6a1de3f4fff19098e2056a85275b7","parents":["d8a24ccee207a1ff24c513fe1c7d3222b3ccd837"],"author":{"name":"Xuan-Son Nguyen","email":"son@huggingface.co","when":"2026-06-13T11:51:20+02:00"},"committer":{"name":"GitHub","email":"noreply@github.com","when":"2026-06-13T11:51:20+02:00"},"message":"server: clean up static assets handling (#24550)\n\n* server: clean up static assets handling\n\n* nits\n\n* simplify file name handling, use static file name everywhere\n\n* cmake/ui : bundle UI assets in an archive\n\n* ui : run prettier on post-build.js\n\n---------\n\nCo-authored-by: Alde Rojas ","tree":"37d64d71e58068bbaf48b4f7cb964c9ba4db55c5"}},{"name":"ui-publish.yml","mode":"0100644","size":2469,"lastCommit":{"hash":"a935fbffe1a3d31509c325c116454ab5d56b2eb8","oid":"a935fbffe1a3d31509c325c116454ab5d56b2eb8","parents":["0badc06ab53a8eb96e01242b92ec1365c4465a2a"],"author":{"name":"Xuan-Son Nguyen","email":"son@huggingface.co","when":"2026-07-10T14:42:17+02:00"},"committer":{"name":"GitHub","email":"noreply@github.com","when":"2026-07-10T14:42:17+02:00"},"message":"server: remove loading.html (#25500)\n\n* server: remove loading.html\n\n* apply ui changes","tree":"340070f349ba519f7f175235c46a8dc46a84d4fd"}},{"name":"ui-self-hosted.yml","mode":"0100644","size":3611,"lastCommit":{"hash":"f7ca93d12c523491e9123ee7ce6a881447c20e52","oid":"f7ca93d12c523491e9123ee7ce6a881447c20e52","parents":["02182fc5b9b0d20c8a4fc15a6b8637dd65d3b537"],"author":{"name":"Aleksander Grygier","email":"aleksander.grygier@gmail.com","when":"2026-06-12T15:53:26+02:00"},"committer":{"name":"GitHub","email":"noreply@github.com","when":"2026-06-12T15:53:26+02:00"},"message":"ui: PWA support (#23871)\n\n* feat: Add basic PWA support and service worker for offline caching\n\n* feat: Vite PWA implementation WIP\n\n* feat: Improve PWA icons generation\n\n* feat: Add PWA workbox to server routes\n\n* feat: Include `version.json` in static assets\n\n* feat: Add HTTP cache headers for PWA static assets\n\n* feat: Update app name for `apple-mobile-web-app-title`\n\n* feat: Implement PWA versioning and automatic update detection\n\n* chore: Update `.gitignore` files\n\n* feat: Splash Screens\n\n* feat: Add dark mode favicon support\n\n* refactor: Cleanup\n\n* fix: Use dark logo for dark splash screens\n\n* refactor: Simplify favicons SVG code\n\n* fix: Adjust caching and polling for reliable service worker updates\n\n* fix: Add missing favicon entry\n\n* fix: Align PWA service worker configuration with SvelteKit build structure\n\n* fix: Replace hashed bundle paths with versioned static paths\n\n* test: Add PWA tests\n\n* ci: Add build output for unit tests\n\n* refactor: Cleanup\n\n* fix: Server build & release versioning\n\n* chore: Update package-lock.json\n\n* chore: Increase PWA cache size\n\n* chore: Update packages\n\n* feat: Update favicons\n\n* refactor: Post-merge fix\n\n* feat: support explicit build version for PWA cache busting\n\n* fix: CI\n\n* feat: Improve PWA Refresh Alert UI\n\n* feat: Add toggleable build version display\n\n* refactor: Cleanup\n\n* feat: Add version mismatch detection and manual app reload\n\n* refactor: replace dynamic imports with static\n\n* refactor: Cleanup\n\n* feat: Add safe space for `pwa-.png` rendered icons\n\n* fix: use relative paths for PWA assets to support base path deployment\n\n* feat: add PWA mode detection via URL query parameter\n\n* feat: Use ?cache=true for SW-cached PWA assets\n\n* refactor: Build process cleanup\n\n* refactor: Decouple PWA versioning and remove ?cache=true workaround\n\n* chore: Update README logo\n\n* feat: Include PWA Assets generation in build script\n\n* refactor: `usePwa` hook for core layout\n\n* fix: Relativize base vite plugin\n\n* fix: remove unnecessary backslash escapes in test regexes\n\n* test: update static asset paths for API Key test\n\n* refactor: Move SvelteKit PWA Options config to constants\n\n* ui: fix update notification never appearing\n\nKeep the PWA hook object intact instead of destructuring needRefreshByStorage,\nwhich freezes the reactive getter. Also exclude loading.html from PWA\nprecache to prevent 404 errors and broken SW installation.","tree":"1daf86fa87d530f15d69823afafd422f524d6ffe"}},{"name":"ui.yml","mode":"0100644","size":4339,"lastCommit":{"hash":"f7ca93d12c523491e9123ee7ce6a881447c20e52","oid":"f7ca93d12c523491e9123ee7ce6a881447c20e52","parents":["02182fc5b9b0d20c8a4fc15a6b8637dd65d3b537"],"author":{"name":"Aleksander Grygier","email":"aleksander.grygier@gmail.com","when":"2026-06-12T15:53:26+02:00"},"committer":{"name":"GitHub","email":"noreply@github.com","when":"2026-06-12T15:53:26+02:00"},"message":"ui: PWA support (#23871)\n\n* feat: Add basic PWA support and service worker for offline caching\n\n* feat: Vite PWA implementation WIP\n\n* feat: Improve PWA icons generation\n\n* feat: Add PWA workbox to server routes\n\n* feat: Include `version.json` in static assets\n\n* feat: Add HTTP cache headers for PWA static assets\n\n* feat: Update app name for `apple-mobile-web-app-title`\n\n* feat: Implement PWA versioning and automatic update detection\n\n* chore: Update `.gitignore` files\n\n* feat: Splash Screens\n\n* feat: Add dark mode favicon support\n\n* refactor: Cleanup\n\n* fix: Use dark logo for dark splash screens\n\n* refactor: Simplify favicons SVG code\n\n* fix: Adjust caching and polling for reliable service worker updates\n\n* fix: Add missing favicon entry\n\n* fix: Align PWA service worker configuration with SvelteKit build structure\n\n* fix: Replace hashed bundle paths with versioned static paths\n\n* test: Add PWA tests\n\n* ci: Add build output for unit tests\n\n* refactor: Cleanup\n\n* fix: Server build & release versioning\n\n* chore: Update package-lock.json\n\n* chore: Increase PWA cache size\n\n* chore: Update packages\n\n* feat: Update favicons\n\n* refactor: Post-merge fix\n\n* feat: support explicit build version for PWA cache busting\n\n* fix: CI\n\n* feat: Improve PWA Refresh Alert UI\n\n* feat: Add toggleable build version display\n\n* refactor: Cleanup\n\n* feat: Add version mismatch detection and manual app reload\n\n* refactor: replace dynamic imports with static\n\n* refactor: Cleanup\n\n* feat: Add safe space for `pwa-.png` rendered icons\n\n* fix: use relative paths for PWA assets to support base path deployment\n\n* feat: add PWA mode detection via URL query parameter\n\n* feat: Use ?cache=true for SW-cached PWA assets\n\n* refactor: Build process cleanup\n\n* refactor: Decouple PWA versioning and remove ?cache=true workaround\n\n* chore: Update README logo\n\n* feat: Include PWA Assets generation in build script\n\n* refactor: `usePwa` hook for core layout\n\n* fix: Relativize base vite plugin\n\n* fix: remove unnecessary backslash escapes in test regexes\n\n* test: update static asset paths for API Key test\n\n* refactor: Move SvelteKit PWA Options config to constants\n\n* ui: fix update notification never appearing\n\nKeep the PWA hook object intact instead of destructuring needRefreshByStorage,\nwhich freezes the reactive getter. Also exclude loading.html from PWA\nprecache to prevent 404 errors and broken SW installation.","tree":"1daf86fa87d530f15d69823afafd422f524d6ffe"}},{"name":"update-ops-docs.yml","mode":"0100644","size":1474,"lastCommit":{"hash":"28123a3937f7d056a4dded43752b216507384373","oid":"28123a3937f7d056a4dded43752b216507384373","parents":["549b9d84330c327e6791fa812a7d60c0cf63572e"],"author":{"name":"Georgi Gerganov","email":"ggerganov@gmail.com","when":"2026-05-25T08:11:19+03:00"},"committer":{"name":"GitHub","email":"noreply@github.com","when":"2026-05-25T08:11:19+03:00"},"message":"ci : move most slim jobs to self-hosted runners (#23619)\n\n* ci : remove tag from build-self-hosted.yml\n\n* ci : slim -> self-hosted\n\n* ci : prevent heavy CPU jobs from running on fast runners\n\n* ci : prevent cmake pkg to run on dedicated fast runners\n\n* ci : try to bump 3.11 -> 3.13\n\n* ci : move lint back to 3.11\n\n* ci : back to 3.11\n\n* ci : add comment about UI jobs\n\n* ci : move python requirements check to CPU runners\n\nthis job is a bit slow for a dedicated \"fast\" runner\n\n* ci : add self-hosted ui workflow\n\n* ci : fix UI naming\n\n* tmp to check if arm64 fast is compatible with all jobs\n\n* revert last commit","tree":"1dd08661bb865152283614cc888549d7516459c4"}},{"name":"winget.yml","mode":"0100644","size":1666,"lastCommit":{"hash":"680a9ae63d60d35c21a0dcd7d3fabdb9c6bfc963","oid":"680a9ae63d60d35c21a0dcd7d3fabdb9c6bfc963","parents":["d8a8beac22d450ebadf175a8ce7b6bf49b66db14"],"author":{"name":"Daniel Bevenius","email":"daniel.bevenius@gmail.com","when":"2026-08-12T14:15:03+02:00"},"committer":{"name":"GitHub","email":"noreply@github.com","when":"2026-08-12T14:15:03+02:00"},"message":"cmake : introduce semantic versioning (#26839)\n\n* cmake : introduce semantic versioning (wip)\n\nThis commit introduces semantic versioning to llama.cpp.\n\n* squash! cmake : introduce semantic versioning (wip)\n\n* cmake : update test-cmake README notes [no ci]\n\n* include libmtmd in output so show its semversioned\n\n* ci : add make-release workflow\n\n* ci : fix build number check in build-cmake-pkg.yml\n\n* examples : remove trailing whitespace\n\n* ci : abort if upstream ggml version does not exist\n\n* ci : extract step contents into scripts\n\n* ci : add GGML_NATIVE=OFF to ubuntu job\n\n* examples : remove CI build information from test-cmake [no ci]\n\nThis commit removes the nightly/release information that I added\npreviously to keep this focused only on using building and installing\nllama.cpp with cmake and being able to quickly verify changes or\ntroubleshoot issues.\n\n* ci : merge scripts into single script\n\n* remove -dev-build_number support\n\nThis commit removes the incremental build number (versioning) support\nthat I added. This was incorrect and we should only use the semver for\nthe version. Releases will be tag a nightly build and package\nmaintainers/managers that build from source can use the tag and it is\ntherefor important that the correct version is reported. So a\nnightly-build will report the semver without the build number. The build\nnumber and commit as availble via cmake and test-cmake has been updated\nto include an example of using them:\n```console\n$ ./build.sh\n[test-cmake] version: 0.1.0, build: 10360 (08c69e381)\n...\n```\n\nRefs: https://github.com/ggml-org/llama.cpp/pull/26839#discussion_r3755836969\n\n* docs: add initial release.md documentation\n\n* cmake : clean-up and add LLAMA_BUILD_IS_DEV option\n\n* ci : remove version input from make-release job\n\n* ci : add LLAMA_BUILD_IS_DEV=OFF to build-cmake-pkg.yml\n\nRefs: https://github.com/danbev/llama.cpp/actions/runs/31576801921/job/94050639145\n\n* docs : update release notes with LLAMA_BUILD_IS_DEV info [no ci]\n\n* ci : add TODO to winget workflow [no ci]\n\n---------\n\nCo-authored-by: Georgi Gerganov ","tree":"bee01f682e7f78eadcaeda93d31f422b103c288f"}}],"lastCommit":{"hash":"9a4dd89d0a6b609e21cd6f43cc4c057f05b356cc","oid":"9a4dd89d0a6b609e21cd6f43cc4c057f05b356cc","parents":["842b1880415d6f508f03b789e5ce70194def7bfd"],"author":{"name":"bri-prism","email":"288398250+bri-prism@users.noreply.github.com","when":"2026-09-25T10:39:48-07:00"},"committer":{"name":"GitHub","email":"noreply@github.com","when":"2026-09-25T10:39:48-07:00"},"message":"release: ship CPU backend variants in the CUDA and HIP bundles (#241) (#271)\n\nThe Windows CUDA x64, Linux CUDA and Windows HIP release jobs build with\nGGML_NATIVE=OFF but without GGML_BACKEND_DL/GGML_CPU_ALL_VARIANTS. With\nnative detection off and no explicit ISA flags, ggml enables AVX/AVX2/FMA/F16C\nby default, so each bundle ships a single AVX2-only ggml-cpu library. On a CPU\nwithout AVX2 the process dies at startup with no message: even\n`llama-server --version` exits silently (#241). The CPU and Vulkan bundles\nalready build every variant and pick one at runtime (Vulkan since #198).\n\nBuild those three bundles with GGML_BACKEND_DL=ON and GGML_CPU_ALL_VARIANTS=ON\nas well. The arm64 CUDA leg already used BACKEND_DL with GGML_CPU=OFF and is\nunchanged.\n\n\nClaude-Session: https://claude.ai/code/session_012Q6ecxozUXFFFLE3N2SMiR\n\nCo-authored-by: Claude Opus 5.5 (1M context) ","tree":"2a993eac47a68f5abf803968e229f6aeb150591f"}}