3 Pull requests merged by 3 people

• [Pytest] Remove obsolete test suite entries

  #18054 merged Jun 12, 2025

• [DOCS] Update installation instruction based ffi refactor

  #18056 merged Jun 12, 2025

• Add support for hamming_window op

  #18036 merged Jun 11, 2025

3 Pull requests opened by 2 people

• [Target] Support CUDA device function calls

  #18055 opened Jun 12, 2025

• [TIR] Phase out ProducerStore, ProducerRealize and Prefetch

  #18057 opened Jun 13, 2025

• [TEST] Move temp files into tempdir

  #18058 opened Jun 13, 2025

2 Issues closed by 2 people

• [Bug] The output of the TVM CPU backend is not identical with those of onnxruntime and TVM CUDA backend

  #18003 closed Jun 13, 2025

• [Docs] Outdated installation documentation

  #18053 closed Jun 12, 2025

3 Unresolved conversations

Sometimes conversations happen on old items that aren’t yet closed. Here is a list of all the Issues and Pull Requests with unresolved conversations.

• [Bug] When compiling using TVM's Relax, the output remains the same regardless of the optimization level set.

  #18048 commented on Jun 11, 2025 • 0 new comments

• [Bug] Runtime Cross Compile Broken After The FFI Refactor

  #18052 commented on Jun 13, 2025 • 0 new comments

• [KVCache] Per Layer Sliding Window

  #17928 commented on Jun 11, 2025 • 0 new comments