LLM Model Notes
Reading an open-weight model card
Separate the artifact from the announcement: revision, total and active parameters, native context, and the license file itself.
Notes
All notes
Every guide on the site, grouped by the question it helps answer. Each note is dated and lists its primary references.
Open models
Evaluation
Reading benchmark results carefully
Define the evaluation contract, inspect uncertainty, and check prompt sensitivity before treating a score as evidence.
Inference
Serving mixture-of-experts models
Why memory follows total parameters while per-token compute follows active ones, and what that means for hardware plans.
Latency, throughput, and context
Measure first-token delay, generation cadence, and capacity under a workload another person can reproduce.
Methods
Measuring tokenizer efficiency for Chinese text
A reproducible way to compare how many tokens a fixed Chinese corpus costs under different tokenizers.
Repeatable model comparisons
Turn a product question into a pinned manifest, paired observations, and a compact result bundle.