MLX vs llama.cpp on Apple Silicon: Which Local LLM Runtime to Actually Use
One is a portable inference engine, the other an array framework that also trains — and that single difference settles almost every practical question.
Latest post · Research
One is a portable inference engine, the other an array framework that also trains — and that single difference settles almost every practical question.
One is a portable inference engine, the other an array framework that also trains — and that single difference settles almost every practical question.
Public benchmarks only shortlist a model; your own cases decide whether to ship it. Here is the dataset, the scorers and the CI gates that catch real drift.
Your agent re-sends the same long prefix on every step and pays to recompute it. Here is the KV mechanism, the ordering rule it forces, and the break-even math.
Decode is memory-bound, so checking several tokens in one pass is nearly free. Here is the mechanism, the arithmetic behind it, and why quality never moves.
Prompting for JSON gets you high-90s reliability; a grammar-driven token mask gets you a guarantee. The mechanism, the real costs, and where it still fails.
Your retriever found the right chunk and buried it at rank 31. A cross-encoder second pass fixes ordering without touching recall, and here is how to size it.
Most teams adopt a specialised store before they need one. An architecture-first look at ANN indexing, filtered search, tenancy and the real thresholds.
Dimension count, sequence length and model size decide your index memory, chunk limits and ingest hours. A framework that prices the retrieval quality you buy.