Deploybase compares GPU cloud and language model API prices, performance benchmarks, and provider market data.
Simple Bench
SimpleBench evaluates everyday reasoning by comparing language models with human performance.
About Simple Bench
SimpleBench is a multiple-choice text benchmark that tests language models on everyday reasoning. It contains over 200 questions covering spatio-temporal reasoning, social intelligence, and linguistic adversarial robustness, described as trick questions. The site reports a non-specialized human baseline of 83.7% from nine participants, compared with 81.9% for Claude Fable.
Visitors can view the latest leaderboard with different views including MCQ, Perf/$, Timeline, and Open-Ended. The site also provides a technical report, a Try Yourself section, a public dataset, code, and a video summary. Benchmark settings are listed as temperature 0.7 and top-p 0.95, except for the o1 series.
It suits researchers, developers, and others interested in how language models compare with human performance on reasoning tasks. The focus on high-school level knowledge makes the questions readable for general readers interested in model evaluation.
Key features
- Over 200 multiple-choice questions on everyday reasoning
- Leaderboard with MCQ, Perf/$, Timeline and Open-Ended views
- Technical report explaining methods and results
- Try Yourself section to attempt benchmark questions
- Public dataset and code available
- Video summary of benchmark updates
Best for: Researchers, developers, and students interested in comparing LLM reasoning with non-specialized human performance.
Good for
- Focuses on everyday reasoning rather than specialized knowledge
- Publishes methods, dataset, and code for review
- Includes human baseline for comparison
- Provides multiple leaderboard views
Watch out for
- Human baseline is based on a small sample of nine participants
Simple Bench at a glance
- Website
- simple-bench.com
- Category
- AI & Machine Learning › AI Data & Analytics
- Type
- Website
- Pricing
- Not confirmed. Check the site
- Platforms
- Web
- Open source
- Not stated
- Language
- English
- Last checked
- Oct 5, 2026 · Online when checked
- Listed
- Oct 5, 2026
Simple Bench alternatives
Tools and sites you could use instead, ranked by how closely they match.
skill-optimizer benchmarks, evaluates, and optimizes skills to check their performance across large language models.
Open source
24 alternatives
AgentBench evaluates large language models acting as agents through a research paper and discussion page.
This article discusses challenges in evaluating language models and interpreting benchmark results.
Context Arena provides a leaderboard for evaluating language models on long-context tasks.
PricePerToken compares API costs for language models from multiple providers.
Popular in AI Data & Analytics
Well-known listings from the same topic.
Dotgetdot.aiForvioforvio.comFluidAI Medicalfluidai.mdApache Zeppelinzeppelin.apache.orgSola Securitysola.securityQuadraticquadratichq.com
Simple Bench FAQ
What is SimpleBench?
SimpleBench is a multiple-choice text benchmark for language models focused on everyday human reasoning. It compares model scores with a non-specialized human baseline.
What topics does SimpleBench cover?
It includes over 200 questions covering spatio-temporal reasoning, social intelligence, and linguistic adversarial robustness, or trick questions.
How do humans and models compare on SimpleBench?
The site reports a human baseline of 83.7% from nine participants, while Claude Fable scored 81.9%. Details on the calculation are in the Human Evaluation section of the report.
Where can I find the SimpleBench dataset and report?
The site links to a public dataset, code, and a full technical report. It also has a Try Yourself section and a video summary.
Is something wrong or out of date?
Tell us and an editor will check it.
Own this website?
Claim this listing
Claim the listing and suggest updates yourself. An editor checks every claim first.