ollama/ollama: v0.33.1
What's Changed MLX: Qwen3.8 Flash Next support cmake: make external compat patches idempotent MLX and llama.cpp update mlxrunner: add structured output support mlxrunner: avoid Metal GPU timeouts when loading models…
Topic - Edition
AI Tools
pytorch/pytorch: v2.14.0-rc9: Pin cython < 3.3.0 for the Windows Triton wheel build (#194931)
Pin cython < 3.3.0 for the Windows Triton wheel build (#194914) Summary Build Triton Windows Wheel (3.15, xpu) fails in setup.py bdist_wheel with exit status 3221225477 (0xC0000005, access violation) — a hard crash, not…
ollama/ollama: v0.33.1-rc1: linux: fix llama.cpp docker build (#18040)
Build context was missing the new cmake common utility.
ollama/ollama: v0.33.1-rc0: MLX: Qwen3.8 Flash Next support (#18032)
MLX: Qwen3.8 Flash Next support review comments
pytorch/pytorch: viable/strict/1787771115
[Testcase Refactoring] Add hw_classification in test_shape_inference.…
pytorch/pytorch: viable/strict/1787762273
[Testcase Refactoring] Add hw_classification in test/fx/test_source_m…
pytorch/pytorch: viable/strict/1787759504
[Testcase Refactoring] Add hw_classification in test/fx/test_subgraph…
vllm-project/vllm: v0.28.0
v0.28.0 Highlights This release features 584 commits from 270 contributors (76 new)! Kimi-K3 performance push: a major optimization effort for Kimi-K3 across the stack — Decode Context Parallel (DCP) support (#50484),…
mlflow/mlflow: v3.15.2
MLflow 3.15.2 is a patch release that includes several major features and improvements. Features: [Evaluation] Support immutable evaluation dataset versions (#24845, @danielseong1) [Evaluation] Add scorer_ensemble…
pytorch/pytorch: trunk/3691693263d2b66a68867e39b7449876844e06cf
Revert "[fix]: derive is_gpu()/device_need_guard() from DeviceInterfa…
pytorch/pytorch: trunk/144ddcdbb75247abac3c9fb01776dd4df1828267: Use C++20 <bit> instead of c10::llvm bit helpers in c10 (#187002)
Description drafted with an AI assistant, quoted below; I've reviewed the change. Replaces the c10::llvm bit helpers with their C++20 <bit> equivalents in c10 and ATen. CachingHostAllocator.h and…
pytorch/pytorch: trunk/96844609ca586bed5757989ac2e620cfb1344ad7
[inductor] lite mode: run reinplacing before decomposing the function…
pytorch/pytorch: viable/strict/1787738484: Re-enable expandable segments for CUDA Graph (#194767) (#194767)
Summary: Previously we temporarily disabled expandable segments for CUDA Graph due to some issues. Now, since the issues were fixed, we re-enable it Test Plan: baseline: f1125979746 QPS: 28.5k peak memory: 97.08%…
pytorch/pytorch: viable/strict/1787736531: [vllm hash update] update the pinned vllm hash (#194672)
This PR is auto-generated nightly by this action. Update the pinned vllm hash. Pull Request resolved: #194672 Approved by: https://github.com/pytorchbot
pytorch/pytorch: trunk/147aaef6de2917b07f797c344bb9ef7b26ec6135: [vllm hash update] update the pinned vllm hash (#194672)
This PR is auto-generated nightly by this action. Update the pinned vllm hash. Pull Request resolved: #194672 Approved by: https://github.com/pytorchbot
pytorch/pytorch: trunk/550ed938e9c427834ec1b9a3e88506e8acc095a1
[fx] Clear fake_mode on a view's base in GraphPickler (#194771) (#194…
pytorch/pytorch: viable/strict/1787727979
Revert "Move c10/util math helper headers into torch/headeronly (#194…
ollama/ollama: v0.33.0
What's Changed Claude Desktop Developers can now easily configure Claude Desktop to seamlessly work with Ollama as a third-party gateway provider. Improved caching Fixed a hang where agent clients that cancel long…
pytorch/pytorch: trunk/14780aaa3ad92d063f64fde1a0e017239164e97b: Move c10/util math helper headers into torch/headeronly (#194673)
Summary Move MathConstants.h, copysign.h, Load.h, BFloat16-math.h, and irange.h from c10/util into torch/headeronly/util under torch::headeronly. Leave the old c10/util paths as one-line forwarding includes, matching…
pytorch/pytorch: trunk/260849cc0ea760acf191de4712381346695cf845
[AOTInductor] Pluggable model-container lifecycle observer interface …
Loading more...
38 items available