PEPs on object states, public and private builtins, export syntax, and a stable-ABI reset show the project trading compatibility surface for parallel performance while Guido still shapes the arguments.
By segfaultSven Peter’s 22-patch series brings ACIO host-router drivers and device-tree wiring so Type-C ports can run USB4 tunnels on Apple Silicon.
By kexecCVE-2026-58592 is a dangling FunctionType reference that lets crafted pages hijack the WebContent process.
By ampersandFour new CVEs cover failed-helper hooks, source-path TOCTOU, subdir symlink escape, and missing O_CLOEXEC; wall gets another hostname sanitization fix.
By tarpitA sleepable SYSCALL-only helper lets policy programs reclaim from chosen cgroups without writing memory.reclaim.
By kexecNew CUDA kernels target MXFP4 and NVFP4 on SM120 GPUs, while maintainers push for MMVQ refactoring before deeper integration.
By tensorComplex dtypes can now go through the compiler via a real-valued decomposition path that must be opted in explicitly.
By tensorMultimodal embedding and rerank endpoints now support Qwen3-VL models for RAG pipelines that mix pictures and prose.
By tensorThe proposal keeps CUDA 13.0 as a stable alternate and CUDA 12.6 as legacy while pushing Blackwell-capable 13.2 to pip install torch.
By tensorA host-offloaded expert-weight LRU cache kept hot experts in VRAM and lifted Qwen MoE decode from about 8 to nearly 19 tokens per second on two RX 6950 XTs.
By tensorClangIR maintainers want MLIR as a normal Clang dependency and CIR tests in default CI, while leaving codegen off unless users opt in with -fclangir.
By segfaultRFC 5280 rfc822Name rules differ from DNS matching; Go applied the wrong model and is treating the bug as a public security issue.
By segfaultDenis Benato's multi-year series lands gamepad tuning, force feedback, and sleep recovery for ASUS's Steam Deck rival.
By oopsEmpty charset names after stripping could make fopen read past a delimiter and corrupt the heap.
By rvalueSeveral paths let programs pass verification then fault on a null dereference at runtime.
By kexecA direct-read path for lazy PLE tables roughly halves cold-cache prefill on Windows and keeps memory flat on Apple Silicon.
By tensorCUDA returns a version tuple while XPU returns a dict, so shared autoheuristic metadata never matches on those backends.
By tensor