configs:
- config_name: models
data_files:
- split: models
path: models.jsonl
tags:
- benchmarks
- evaluation
- decision-models
- leaderboard
Decision Model Benchmark Usage
Which evaluation benchmarks labs use to evaluate their decision models (single-pass,
zero-output-token classifiers/raters — routing, triage, moderation, reranking, LLM-judge,
visual inspection — not chat LLMs). Sister dataset to
SaylorTwift/llm-benchmark-usage,
which covers LLM/agent evaluation suites; benchmark names are canonicalized against
that dataset so overlapping benchmarks (e.g. AI2D, MMBench, BoolQ) share one spelling
across both.
Hand-built from papers, technical reports, system cards, model cards, and blog posts.
from datasets import load_dataset
models = load_dataset("SaylorTwift/decision-model-benchmark-usage", "models")["models"]
models
One row per model, fully self-contained.
| column | type | description |
|---|---|---|
model_id |
string | Hugging Face repo id for open-weight models (e.g. LiquidAI/d1-3B), or a plain slug for closed/API models |
lab |
string | Organization/lab that released the model |
release_date |
timestamp | Release date. HF repo creation date for open-weight models (a proxy, not always the exact announcement date); hand-researched announcement date for closed models |
source |
string | Link to the paper/report/card/blog this model's evaluation suite was extracted from |
benchmarks |
list | The evaluation suite: {name, category} objects. Names are canonicalized (e.g. always CVBench, never CV-Bench); genuinely distinct variants are kept separate (SQuAD vs SQuAD 2.0). Categories are per-source free-text groupings, not a controlled vocabulary |
Richer per-source metadata (paper type, title, notes, confidence flags) lives in a local
sources table and is not published as of this schema version; it may move to a separate
Hub dataset later.