ReleaseWatch – lokaler Release-Radar

Automatisch gesammelt, nicht redaktionell verifiziert. Sicherheitsmarker sind nur Stichworttreffer.

uv Releases · 2026-10-14T05:15:30Z · Release / Änderung

0.13.1

Release Notes Released on 2026-10-14. Enhancements Ignore Requires-Python upper bounds in uv pip check and uv pip install --strict , consistent with resolution ( #22500 ) Support Linux SPARC by falling back when filesystem cloning and detailed cache accounting are unavailable ( #22463 ) Report missing trusted hashes in resolution errors ( #22241 ) Python Add CPython 3.16.0a1 ( #22527 ) Preview features Allow uv tool upgrade to upgrade installed tools when tool-install-locks is enabled ( #22498 ) Add preliminary support for parsing PEP 825 variant labels in wheel filenames with wheel-variants ( #22039 ) Performance Reduce uv's binary size ( #22429 , #22430 , #22432 ) Speed up Simple API parsing ( #22429 ) Reduce package metadata parsing overhead ( #22239 ) Bug fixes Prevent tool and managed Python uninstalls from deleting unrelated files beside their storage directories ( #22495 ) Restore support for source distributions containing hard links or full-length tar user or group names ( #22516 , #22522 ) Preserve device files and FIFOs used as uv pip compile or uv export output destinations ( #22496 ) Fix sys._base_executable detection for environments created from symlinked interpreter

uv Releases · 2026-10-09T19:49:55Z · Release / Änderung

0.13.0

Release Notes Released on 2026-10-09. uv 0.13.0 makes Python 3.15 the default stable Python version. We've also included several breaking changes to improve correctness, performance, and compatibility, described below. We expect most users to be able to upgrade without making changes. While not a breaking change, this release also updates the format of many of uv's cache entries to improve performance. uv may download or rebuild dependencies after upgrading , because some cached entries from earlier versions cannot be reused. Multiple versions of uv can still safely share the same cache directory. There are no breaking changes to the configuration of the uv build backend . If your [build-system] table includes an upper bound on uv_build , update it to allow uv_build 0.13, e.g., uv_build>=0.13.0,<0.14 . Breaking changes Use Python 3.15 as the default stable version The default stable Python version has changed from 3.14 to 3.15. This affects Python downloads when no version is requested or pinned, e.g., when running uv python install . uv continues to use compatible Python installations that are already present. For example, uv venv can still use an installed Python 3.14. If no suit

uv Releases · 2026-10-08T20:06:20Z · Release / Änderung

0.12.24

Release Notes Released on 2026-10-08. Enhancements Remove orphaned temporary build environments with uv cache prune ( #22171 ) Accept PEP 508 marker operators directly before grouped expressions ( #22309 ) Reject malformed requirements-file options instead of partially parsing or ignoring them ( #22317 ) Show underlying filesystem and registry errors when managed Python uninstallation fails ( #22362 ) Identify the invalid source URL in Python mirror errors ( #22364 ) Preview features Display preferred advisory IDs in uv audit reports, prioritizing PYSEC, GHSA, then CVE identifiers ( #22292 ) Configuration Support custom installation mirrors for GraalPy ( #22269 ) Support custom installation mirrors for Pyodide ( #22271 ) Allow UV_NO_CACHE=false to override no-cache = true in configuration ( #22324 ) Report more precise error locations for invalid trusted-host ports and preview-feature list entries ( #22144 ) Performance Speed up later commands after creating an environment by warming its interpreter cache ( #21304 ) Reduce code-signature verification work for ARM64 macOS releases with 16 KiB signature pages ( #22246 ) Enforce resource limits when parsing package indexes and --find-

uv Releases · 2026-10-27T00:47:49Z · Release / Änderung

0.13.6

Release Notes Released on 2026-10-26. Enhancements Accept an empty UV_RUN_RECURSION_DEPTH as zero and report invalid values consistently ( #22814 ) Explain how to add project metadata when uv init finds a pyproject.toml without a [project] table ( #22916 ) Python Add CPython 3.11.18, 3.12.16, 3.13.18, 3.14.10, and 3.15.2 ( #22938 ) Preview features Detect NVIDIA GPU capabilities when selecting compatible wheel variants with wheel-variants ( #22690 ) Nest each module's owners in an object in uv workspace metadata output ( #22638 ) Report the unsupported properties required by an incompatible wheel variant ( #22719 ) Performance Reduce peak memory use when resolving large workspaces ( #22860 , #22863 , #22864 , #22866 , #22870 ) Reduce peak memory use when locking or upgrading large projects ( #22833 , #22834 ) Speed up redirected output from uv export and uv pip compile ( #22840 ) Reuse cached or active environments when running PEP 723 scripts without dependencies ( #21458 ) Reduce parsing overhead for build dependency and build backend settings ( #22848 , #22871 , #22879 , #22884 ) Reduce parsing overhead for project sources and workspace references ( #22880 , #22883 ) Reduce pars

uv Releases · 2026-10-02T00:20:42Z · Release / Änderung

0.12.22

Release Notes Released on 2026-10-01. Python Add CPython 3.10.22, 3.11.17, 3.12.15, 3.13.16, and 3.14.8 ( #22147 ) Enhancements Accept uppercase release suffixes in wheel platform tags ( #22113 ) Record workspace-member default groups in lockfiles ( #22010 , #22103 ) Record workspace-member dependency-group Python requirements in lockfiles ( #22044 , #22103 ) Record default groups for non-project workspace roots in lockfiles ( #22104 ) Record dependency-group Python requirements for non-project workspace roots in lockfiles ( #22104 ) Format URLs and paths consistently in CLI messages ( #21937 ) Hide the unsupported --offline option from uv publish help ( #22124 ) Preview features Honor --no-default-groups in uv audit ( #22090 ) Report a clear error when uv audit or uv tool audit runs offline and hide the unsupported option from help ( #22114 ) Configuration Add UV_PYTHON_ARCH to select an interpreter architecture independently of its Python version ( #22098 ) Performance Reduce uv's binary size by compressing embedded Python download metadata ( #22126 ) Bug fixes Verify unchanged requirements against existing lockfile hashes when relocking ( #22083 ) Honor dependency-group Python r

uv Releases · 2026-10-20T01:15:17Z · Release / Änderung

0.13.3

Release Notes Released on 2026-10-20. Performance Avoid severe uv export slowdowns in dependency graphs with extras ( #22750 ) Install uv 0.13.3 Install prebuilt binaries via shell script curl --proto ' =https ' --tlsv1.2 -LsSf https://releases.astral.sh/github/uv/releases/download/0.13.3/uv-installer.sh | sh Install prebuilt binaries via powershell script powershell -ExecutionPolicy Bypass -c " irm https://releases.astral.sh/github/uv/releases/download/0.13.3/uv-installer.ps1 | iex " Download uv 0.13.3 File Platform Checksum uv-aarch64-apple-darwin.tar.gz Apple Silicon macOS checksum uv-x86_64-apple-darwin.tar.gz Intel macOS checksum uv-aarch64-pc-windows-msvc.zip ARM64 Windows checksum uv-i686-pc-windows-msvc.zip x86 Windows checksum uv-x86_64-pc-windows-msvc.zip x64 Windows checksum uv-aarch64-unknown-linux-gnu.tar.gz ARM64 Linux checksum uv-i686-unknown-linux-gnu.tar.gz x86 Linux checksum uv-powerpc64le-unknown-linux-gnu.tar.gz PPC64LE Linux checksum uv-riscv64gc-unknown-linux-gnu.tar.gz RISCV Linux checksum uv-s390x-unknown-linux-gnu.tar.gz S390x Linux checksum uv-x86_64-unknown-linux-gnu.tar.gz x64 Linux checksum uv-armv7-unknown-linux-gnueabihf.tar.gz ARMv7 Linux checksum uv

uv Releases · 2026-10-20T16:53:15Z · Release / Änderung

0.13.4

Release Notes Released on 2026-10-20. This release addresses GHSA-hq85-5cpj-mr62 , which involves a root CA whose intermediate on the Web PKI have been compromised. Users are encouraged to upgrade immediately, as well as confirm that all other software that uses bundled CAs is also properly updated. Preview features Validate PEP 825 variant metadata and treat package versions with invalid metadata as unavailable under wheel-variants ( #22512 ) Other changes Update the bundled Mozilla root certificates via webpki-root-certs 1.0.10 ( #22760 ) Install uv 0.13.4 Install prebuilt binaries via shell script curl --proto ' =https ' --tlsv1.2 -LsSf https://releases.astral.sh/github/uv/releases/download/0.13.4/uv-installer.sh | sh Install prebuilt binaries via powershell script powershell -ExecutionPolicy Bypass -c " irm https://releases.astral.sh/github/uv/releases/download/0.13.4/uv-installer.ps1 | iex " Download uv 0.13.4 File Platform Checksum uv-aarch64-apple-darwin.tar.gz Apple Silicon macOS checksum uv-x86_64-apple-darwin.tar.gz Intel macOS checksum uv-aarch64-pc-windows-msvc.zip ARM64 Windows checksum uv-i686-pc-windows-msvc.zip x86 Windows checksum uv-x86_64-pc-windows-msvc.zip x64

uv Releases · 2026-10-03T17:34:11Z · Release / Änderung

0.12.23

Release Notes Released on 2026-10-03. Python Add CPython 3.15.0rc3 ( #22164 ) Preview features Sync from uv.lock without a workspace manifest using uv sync --frozen with frozen-lockfile ( #22018 ) Export from uv.lock without a workspace manifest using uv export --frozen with frozen-lockfile ( #22007 ) Inspect dependency trees from uv.lock without a workspace manifest using uv tree --frozen with frozen-lockfile ( #22016 ) Inspect workspace metadata and optionally sync its environment from uv.lock without a workspace manifest using uv workspace metadata --frozen with frozen-lockfile ( #22017 , #22018 ) Bug fixes Reject alternate sources for workspace members across conflicting dependency selections, avoiding lockfiles that cannot be installed ( #22153 ) Allow x86-64 Python interpreters running under emulation on Windows ARM64 to install compatible win_amd64 wheels instead of building from source ( #22099 ) Install uv 0.12.23 Install prebuilt binaries via shell script curl --proto ' =https ' --tlsv1.2 -LsSf https://releases.astral.sh/github/uv/releases/download/0.12.23/uv-installer.sh | sh Install prebuilt binaries via powershell script powershell -ExecutionPolicy Bypass -c " irm http

uv Releases · 2026-10-19T21:02:37Z · Release / Änderung

0.13.2

Release Notes Released on 2026-10-19. Enhancements Report all installed packages in uv workspace metadata , including packages outside the selected resolution ( #22395 ) Include supported environment restrictions when exporting pylock.toml ( #22596 ) Retry dropped GET and HEAD requests when fetching package and index data ( #22672 ) Preserve encoding declarations in wheel data scripts that require shell launchers ( #22526 ) Show causes and hints when uv python find --script cannot discover an interpreter, including with -q ( #22549 ) Print valid ksh activation instructions and support non-relocatable environments under OpenBSD-derived ksh ( #22538 , #22572 ) Allow PEP 508 direct URLs, including Mercurial and extensionless archive URLs, in uv-build's [build-system].requires validation ( #22577 ) Python Add PyPy 3.11.17 and 3.12.15 ( #22578 ) Preview features Fetch and parse PEP 825 index-level variant metadata with wheel-variants ( #22509 , #22511 ) Performance Avoid exponential work when simplifying, traversing, and exporting complex dependency markers ( #22608 , #22615 , #22651 , #22674 ) Reduce the time and memory used to parse and store complex dependency markers ( #22675 , #226

uv Releases · 2026-10-22T01:35:54Z · Sicherheitsbezug prüfen

0.13.5

Release Notes Released on 2026-10-21. Following uv 0.13.4 , which removed the affected eMudhra/emSign root from uv's own certificate store, this release updates the latest managed CPython builds for 3.11 and newer to python-build-standalone 20261021 , including the CPython 3.15.1 , 3.14.9 , and 3.13.17 security patch releases. These builds include pip 26.2.2 both preinstalled in the managed interpreter and bundled with ensurepip , which installs pip by default when you run python -m venv . Both copies use certifi 2026.10.19, which removes the same affected root. After upgrading uv, run uv python upgrade to install the updated builds. Virtual environments pinned to an older Python patch version do not switch to the updated interpreter automatically. Upgrade pip separately in existing virtual environments. Please note upstream CPython 3.9 and 3.10 still bundle pip releases whose certificate bundles trust the affected eMudhra/emSign root. We recommend migrating off of those versions and will no longer be releasing Docker images with uv installed for EOL Python versions. Enhancements Report incompatible self-dependencies directly in resolution errors ( #21845 ) Python Add CPython 3.15.

Ollama Releases · 2026-10-27T21:41:00Z · Release / Änderung

v0.40.12

Liquid AI LFM3 LFM3 is Liquid AI's new family of small models built for local agents, tool calling, and structured output, on Liquid's new tri-hybrid architecture. The family includes: lfm3 : a 1.7B text model with a 256K-token context window lfm3-vl : adds image understanding lfm3-audio : speech recognition, translation, and text-to-speech in 18 languages lfm3-omni : text, images, and audio in one model ollama run lfm3 d1 decision models Liquid AI's d1 models now come in a 2B size, built on LFM3. d1 makes decisions about text and images, and d1-omni adds audio. ollama run d1:2b " I was charged twice for order 4417. " --ask " Is this a billing issue? " Note: lfm3 and d1 2b are supported on Apple hardware in this release. Support for Linux/Windows coming soon. What's Changed Faster prompt processing on Apple silicon for Gemma 4, Flint and Laguna models Speculative decoding now supports DFlash2 draft models on MLX Gemma 4 keeps more detail in small images ollama launch now works with OpenCode v2 Fixed tool-only assistant turns from Kimi CLI Updated MLX and llama.cpp Full Changelog : v0.40.11...v0.40.12

Ollama Releases · 2026-10-27T15:47:18Z · Release / Änderung

v0.40.12-rc1

mlx: add LFM3 and d1 models ( #19333 )

Ollama Releases · 2026-10-25T18:06:57Z · Release / Änderung

v0.40.11

Speech Ollama can now generate speech on macOS. ollama run plays text or a chat model's reply aloud as the audio is generated, and apps can stream audio through the OpenAI-compatible /v1/audio/speech endpoint. ollama run phonon " Hello from Ollama. " d3 decision models d3 brings vLLM Semantic Router's decision models to Ollama. Give a model text, structured state, or an image, and ask it to classify, answer yes or no, or score options with probabilities. The family includes 2B, 4B, 9B, and 27B sizes. ollama run d3 " A customer says their card was charged twice. " --ask " Is this a billing issue? " New Contributors @SynapShift made their first contribution in #19173 Full Changelog : v0.40.10...v0.40.11

Ollama Releases · 2026-10-23T23:15:08Z · Release / Änderung

v0.40.11-rc0: voice: add voice catalog for Phonon (#19242)

mlxrunner: import and list Phonon catalog voices docs: document Phonon voices

Ollama Releases · 2026-10-28T22:20:04Z · Release / Änderung

v0.40.13-rc1: llama.cpp: workaround Metal unary crash (#19372)

This temporary patch works around a regression on older versions of MacOS ggml_metal_library_compile_all: failed to build 'unary' library

Ollama Releases · 2026-10-28T17:37:56Z · Release / Änderung

v0.40.13-rc0

llama.cpp: version bump b11963 ( #19360 )

Ollama Releases · 2026-10-29T02:54:11Z · Release / Änderung

v0.40.13

LFM3 on Linux and Windows Liquid AI's LFM3 models, including text-to-speech in lfm3-audio and lfm3-omni , now run on Linux and Windows as well as Apple silicon. Full Changelog : v0.40.12...v0.40.13

Ollama Releases · 2026-10-26T23:10:42Z · Release / Änderung

v0.40.12-rc0

llama.cpp: version bump b11901 ( #19311 )

Ollama Releases · 2026-10-23T14:12:11Z · Release / Änderung

v0.40.10-rc0: Update gemma renderers (#19241)

renderers: add gemma4-nemotron renderer Add a Gemma 4 renderer variant that renders assistant content before its tool calls rather than after the tool responses. Models opt in with RENDERER gemma4-nemotron; gemma4 auto-detection is unchanged. Adds the corresponding chat template as a test fixture for Jinja parity checks. mlx: support Gemma 4 targets with DFlash drafts Load a draft-owned output head when its config requires one, while preserving the target-head path used by existing DFlash models. Expose Gemma 4 hidden-state taps and raw logits for speculative decoding. Report safetensors target precision independently of quantized draft weights.

Ollama Releases · 2026-10-23T17:17:19Z · Release / Änderung

v0.40.10

NVIDIA Nemotron 3.5 Bolt E4B Nemotron 3.5 Bolt E4B is NVIDIA’s smaller, Gemma 4-based vision-language model for fast local conversations, image understanding, and tool use. It supports a 128K-token context window. The model includes a DFlash draft for speculative decoding. ollama run nemotron-3.5-bolt:e4b Full Changelog : v0.40.9...v0.40.10