PageIndex

PageIndex replaces vector search with a tree index an LLM navigates. How it really works, why the headline benchmark is 90.7% strict, and how to run it against a local model.

TensorFold vs MTPLX vs mlx-dspark

Three projects claim faster local decoding with byte-identical output. What exactness means, the author-reported numbers, the MLX version that breaks it, and which to run.

DeepSeek Harness: The Open-Source Agent Layer Nobody Owns Yet

DeepSeek shipped the middle layer of the agent stack under MIT and told everyone to take it apart. Here is what dsh actually is, what the GitHub and npm APIs really say about its adoption, and why a free harness arrived three days before V4 output tokens got 4.5 times more expensive at peak.