3 articles tagged
Sentence Transformers 6.0 published 18 August 2026 adds ColBERT-style late interaction. The gain is about one NDCG point on NanoBEIR. The cost is 311.5 MB where a dense index used 7.5 MB.
PyTorch 2.13 landed FlexAttention on the Mac's Metal backend, with a documented ~12x speedup over SDPA on sparse masks. Here is what the numbers mean, how to write a mask in two lines, and where MLX still wins.
PyTorch 2.13 lands FlexAttention on Apple Silicon with Metal kernels for sparse prefill and decode, up to 12x faster than SDPA. How the API works, what the numbers show, and when block-sparse attention pays off on a Mac.
