Django
Django Benchmark CI Adds Ubuntu Results
The django-asv project quietly extended its benchmark matrix to ubuntu-latest, expanding coverage with a single commit.
read --wire →
$ tail -f topics/python.log
Python news, every morning.
Daily updates from across the Python ecosystem - Django, FastAPI, PyTorch, Hugging Face, and the data-and-AI stack.
34 updates across 5 projects this week.
Your week in Python, every Monday.
One calm review of what shipped across Python - the commits, releases, and security advisories that matter. Every Monday, with security advisories same-day. Free, unsubscribe in one click.
We'll start you on the top Python projects - refine anytime. · Read a sample issue →
Recent wires
FastAPI & Pydantic
CI CACHE GOES DARK ACROSS FASTAPI AND SQLMODEL
A routine bump to setup-uv just disabled CI caching by default for three event types in both FastAPI and SQLModel.
read --wire →
Hugging Face
TRL UNBLOCKS MOE LORA AS PEFT PATCHES AUTOCAST
The MoE LoRA CI failures that have been plaguing TRL for days are gone, and the fix came from upstream.
read --wire →
PyTorch
EXECUTORCH CUTS GRAPH REPLAYS, ARM KILLS NO-OP RESIZES
Two executorch optimizations landed overnight that could shave real runtime on device, while PyTorch's CI finally silences the Node-20 roar.
read --wire →
GENAI SHIPS 2.20.0 WHILE API-CORE PATCHES A YANKED RELEASE
If you pulled the latest google-cloud-python, check your Firestore calls: the api-core path_template change broke gRPC requests, and the fix is landing now.
read --wire →
Django
One Commit, One Result: django-asv Adds Ubuntu Support
A single bot commit proves the value of continuous benchmarking as django-asv now tracks performance on an additional platform.
read --wire →
FastAPI & Pydantic
PYDANTIC FIXES SERIALIZER EXCLUSION BUG
Pydantic's most requested validation fix just landed, and it's exactly what your custom serializers have been waiting for.
read --wire →
Hugging Face
Transformers MoE loss cut 99.7% as TRL gains tool calling
A single PR eliminates a 128k-token memory bomb in MoE training while TRL's distillation trainer learns to call tools.
read --wire →
JAX restores fast multi-dimensional IRFFT on GPU
A JAX patch cuts multi-dimensional IRFFT time on GPU by roughly 1.7x, restoring the single C2R transform after years of slower decomposition.
read --wire →
Django
DJANGOPROJECT.COM SHORES UP CI TOOLING WITH ACTION UPGRADES AND PRE-COMMIT AUTOMATION
Django's website infrastructure tightened its release pipeline overnight with setup-python v7 and a new pre-commit ecosystem strategy.
read --wire →
Hugging Face
Transformers linter hunts nn.Buffer misuse as kernels ships hotfix
transformers-mlinter is tightening model internals while kernels drops a quick patch, but the real action is a quiet alignment fix in trl.
read --wire →
PyTorch
FSDP2 CRASH FIXED, EXECUTORCH BUILD PIPELINE REPAIRED AFTER COMMIT PROMOTION BUG
PyTorch's distributed training guard catches an AttributeError that was silently breaking gradient accumulation in sharded models, while ExecuTorch's nightly builds restart after a scheduling bug locked out half of all commits.
read --wire →
JAX SPEEDS UP SNAKE, DROPS ROCM WHEEL LOOKUP
JAX's planar_snake is now up to 10% faster for power-of-two tiles, and the team is killing a painfully slow ROCm wheel discovery path.
read --wire →
Django
Django fixes release note, clears default ordering confusion
A single-word fix in a release note clarifies a behavior change that could have tripped up upgrade teams.
read --wire →
Hugging Face
Harbor-hf sinks broken Sandbox cleanup
Harbor-hf shipped five fixes overnight that close the gaps where stuck Sandboxes, terminal jobs, and invalid campaign results could quietly block recovery.
read --wire →
PyTorch
PyTorch's Caffe2 cleanup hits a wall as internal dependencies force three reverts
PyTorch's attempt to shed its Caffe2 legacy just got blocked by internal dependencies, and the resulting reverts ripple across the cleanup effort.
read --wire →
JAX Renames Mosaic GPU Parameters
Google's JAX team is renaming a core Mosaic GPU attribute, and it will break your cross-process parameter code.
read --wire →
Django
Django ASV Gains RelatedIn Benchmark
A new benchmark lands in django-asv, zeroing in on the RelatedIn lookup that's been haunting issue #37237.
read --wire →
Hugging Face
LEROBOT CUTS FASTWAM OVERHEAD WITH COMPILED INFERENCE
LeRobot's FastWAM policy gets an opt-in compiled path that slashes Python and CUDA kernel-launch overhead on cached action inference.
read --wire →
PyTorch
Pallas HBM Windows Open on Any Tensor Axis
Pallas stagers just gained the ability to align HBM windows on any tensor axis, not just the last one, and a critical autotuning crash is fixed.
read --wire →