7 Pull requests merged by 4 people

• spm : remove

  #1135 merged Mar 8, 2025

• sync : whisper.cpp

  #1137 merged Mar 8, 2025

• sync : llama.cpp

  #1134 merged Mar 7, 2025

• ggml_compute_forward_concat() for arbitrary tensor type

  #1118 merged Mar 7, 2025

• pkg-config: Use CMake install paths for lib, include

  #1133 merged Mar 6, 2025

• sync : llama.cpp

  #1132 merged Mar 4, 2025

• cuda: unary ops - perform the final computation as a float, and de-duplicate the code

  #1130 merged Mar 3, 2025

1 Issue closed by 1 person

• No viable overloaded '='

  #1136 closed Mar 8, 2025

1 Issue opened by 1 person

• How to use intermediate buffer during the build graph, so that graph_compute accounts it during inference

  #1131 opened Mar 4, 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.

• Tests, examples fail to build with GGML_BACKEND_DL=ON

  #1120 commented on Mar 3, 2025 • 0 new comments

• ggml : add FFT operator

  #1049 commented on Mar 7, 2025 • 0 new comments

• ggml : add `ggml_fft` and `ggml_ifft` operator

  #1105 commented on Mar 7, 2025 • 0 new comments