# Glossary

- **Benchmark**: A standard test used to compare AI models on the same task. Like an exam that every model takes under the same rules, so the scores can be compared. A benchmark only tells you about the task it tests — a model that wins at coding may be average at reading handwriting.
- **Leaderboard**: A ranked table of models' scores on a benchmark. Usually maintained by the benchmark's authors. Check its date: models change every few weeks.
- **LLM**: Large language model — the kind of AI behind chat assistants. A model trained on huge amounts of text that can write, answer questions, summarize and follow instructions.
- **OCR**: Reading text from images (photos, scans). Optical character recognition. Here usually done by vision-language models that look at the photo and write out the text.
- **ASR**: Speech-to-text: turning spoken audio into written text. Automatic speech recognition, also called transcription.
- **TTS**: Text-to-speech: turning text into spoken audio.
- **WER**: Word error rate: share of words transcribed wrong. Lower is better. Counts wrong, missing and extra words against the correct text. 10% means roughly 1 word in 10 is wrong. It can exceed 100% if the model adds a lot of text that wasn't said.
- **CER**: Character error rate: share of characters read wrong. Lower is better. Like WER but per letter. 0.02 (or 2%) means about 2 wrong characters in 100.
- **DER**: Diarization error rate: share of time given to the wrong speaker. Lower is better.
- **Diarization**: Working out who is speaking when in a recording.
- **Embeddings**: Turning text into lists of numbers so that similar meanings end up close together. This is what powers search and 'find similar' features: the question and the documents are embedded, and the closest documents are returned.
- **Accuracy**: Share of answers that were correct. Higher is better.
- **Latency**: How long you wait for an answer, in seconds. Lower is better. Often split into time to first token (when the answer starts appearing) and total time (when it is finished).
- **Token**: A piece of text (roughly ¾ of a word) — the unit models read, write and bill by.
- **Reasoning**: Hidden 'thinking' a model does before answering. Reasoning models write internal steps (reasoning tokens) before the answer. It can improve hard tasks but costs time and money, because those tokens are billed too.
- **Elo**: A rating from head-to-head votes, like in chess. Higher is better. Used by arena leaderboards: people compare two anonymous answers and pick the better one. A 100-point gap means the higher model wins roughly 64% of matchups.
- **AUC**: How well a classifier separates two groups: 1.0 is perfect, 0.5 is a coin flip.
- **Recall**: Share of the things you were looking for that were actually found. Higher is better.
- **Precision**: Share of the things flagged that were actually correct. Higher is better.
- **False positive**: A false alarm: something harmless flagged as a problem. Lower is better.
- **pass@1**: Share of tasks solved on the first try. Higher is better.
- **Platform**: The service you call the model through (e.g. an API provider). The same model can behave differently on different platforms.
- **Open-weights model**: A model whose files are published, so anyone can run it on their own hardware.
- **Saturated**: A benchmark where top models score near the maximum, so it no longer separates them well.
- **Contamination**: When test questions leaked into a model's training data, inflating its score.
