Files
whisper.cpp/ggml
SXX bfe3dc63c1 ggml-cpu: support BF16/FP16/FP32 K tails in tinyBLAS on x86 (llama/29806)
* ggml-cpu: vectorize BF16 K tails in tinyBLAS

* tests: Skip tinyBLAS when use_ref is enabled so CPU tests compare against the vec_dot path.

* ggml-cpu: vectorize tinyBLAS F16/F32 tails
2026-10-06 10:38:20 +03:00
..