Popular repositories Loading

1. mlx-benchmarks mlx-benchmarks Public

   Systematic MLX inference benchmarks on Apple Silicon. 290 data points across 10 models (8B to 1T), 5 quantisation levels, 5 topologies. Cross-platform comparison with NVIDIA.

   Python 3

2. australian-english-llm australian-english-llm Public

   Falsifiable, stage-gated methodology for resource-disciplined LLM specialisation — worked example: adapting a generic LLM to believable (non-cliché) Australian English. Work in progress (v0.1).

   Python 1

3. agent-callisto2 agent-callisto2 Public

   Forked from smiley/hyper-robco

   Adds Fallout terminal sound effects to your Hyper terminal. Prepare for the Future, today!

   JavaScript

4. GenomicWeightsThesis GenomicWeightsThesis Public

   Python

5. gwt-visualisation gwt-visualisation Public

   A visualisation of the results from my GWT

   TypeScript

6. wcer wcer Public

   WCER — Workload-Conditioned Expert Residency: run a Mixture-of-Experts model using only the experts your workload actually uses (memory + time-to-first-token savings, behavior-preserving). Validate…

   Python