# Human Behavior Bench V1 (Kaggle) (Human Behavior Bench)

> Predicts response distributions for specified groups of U.S. adults on closed-form questions about behavior and preferences.

Canonical page: https://benchlm.ai/benchmarks/human-behavior-bench

- Category: [Reasoning](/reasoning)
- Last updated: 2026-10-20 source capture

Evaluation results published by [Kaggle Benchmarks](https://www.kaggle.com/benchmarks/rehearsals/human-behavior-bench). undefined

Human Behavior Bench asks a model to predict how a specified group of U.S. adults answered a closed-form survey question. The V1 dataset has 1,199 tasks: 239 public and 960 private.

We show the Kaggle Combined Score as a percentage. Agreement is one minus total variation distance between the predicted and observed answer distributions. Public and private scores remain separate in the source snapshot.

This board contains Kaggle model evaluations. Rehearsals Engram uses a separate in-house simulation harness on the owner website and is not merged into this table. Survey agreement does not establish individual behavior or future purchasing accuracy.

These are published source configurations. Compare the task version and disclosed settings before treating a score difference as a model difference.

- [Kaggle leaderboard and methodology](https://www.kaggle.com/benchmarks/rehearsals/human-behavior-bench)
- [Published leaderboard data](https://www.kaggle.com/api/v1/benchmarks/rehearsals/human-behavior-bench/leaderboard)
- [Rehearsals methodology and separate Engram report](https://www.runrehearsals.com/human-behavior-bench)
- [Apache 2.0 license shown by Kaggle](https://www.apache.org/licenses/LICENSE-2.0)

## About Human Behavior Bench

- Year: 2026
- Tasks: 1,199 tasks; 239 public and 960 private
- Format: Published Kaggle leaderboard aggregates
- Paper: [Human Behavior Bench on Kaggle](https://www.kaggle.com/benchmarks/rehearsals/human-behavior-bench)

The combined agreement score is one minus total variation distance, shown on a 0–100 percentage scale. The public and private task splits remain available in the source snapshot. We keep the Kaggle harness separate from the owner's Engram simulation report.

Human Behavior Bench is currently displayed on BenchLM for reference, but it is excluded from the weighted scoring formula.

## Leaderboard (11 configurations)

| Rank | Model | Configuration | Creator | Score |
|------|-------|---------------|---------|-------|
| 1 | [Gemini 4 Argon](https://www.kaggle.com/benchmarks/rehearsals/human-behavior-bench) | Task version 1; Evaluated 2026-10-13 | — | 80.6% |
| 2 | [Claude Opus 5](https://www.kaggle.com/benchmarks/rehearsals/human-behavior-bench) | Task version 1; Evaluated 2026-09-30 | — | 80.2% |
| 3 | [Gemini 3.7 Flash](https://www.kaggle.com/benchmarks/rehearsals/human-behavior-bench) | Task version 1; Evaluated 2026-10-02 | — | 79.7% |
| 4 | [GPT-5.6 Sol](https://www.kaggle.com/benchmarks/rehearsals/human-behavior-bench) | Task version 1; Evaluated 2026-10-05 | — | 79.5% |
| 5 | [Claude Opus 5.5](https://www.kaggle.com/benchmarks/rehearsals/human-behavior-bench) | Task version 1; Evaluated 2026-10-05 | — | 79.3% |
| 6 | [Gemini 3.8 Flash](https://www.kaggle.com/benchmarks/rehearsals/human-behavior-bench) | Task version 1; Evaluated 2026-10-05 | — | 79.2% |
| 7 | [Gemini 3 Flash Preview](https://www.kaggle.com/benchmarks/rehearsals/human-behavior-bench) | Task version 1; Evaluated 2026-10-02 | — | 78.5% |
| 8 | [GPT-5.6 Terra](https://www.kaggle.com/benchmarks/rehearsals/human-behavior-bench) | Task version 1; Evaluated 2026-09-30 | — | 77.8% |
| 9 | [Gemini 3.5 Flash-Lite](https://www.kaggle.com/benchmarks/rehearsals/human-behavior-bench) | Task version 1; Evaluated 2026-09-30 | — | 75.9% |
| 10 | [Grok 4.7](https://www.kaggle.com/benchmarks/rehearsals/human-behavior-bench) | Task version 1; Evaluated 2026-10-16 | — | 75.2% |
| 11 | [Grok 4.20 (Non-Reasoning)](https://www.kaggle.com/benchmarks/rehearsals/human-behavior-bench) | Task version 1; Evaluated 2026-10-05 | — | 73.5% |

## FAQ

### What does Human Behavior Bench measure?

Predicts response distributions for specified groups of U.S. adults on closed-form questions about behavior and preferences.

### Which source row has the highest reported Human Behavior Bench value?

Gemini 4 Argon has the highest reported Human Behavior Bench value at 80.6%. These are published source configurations. Compare the task version and disclosed settings before treating a score difference as a model difference.

### How many models are evaluated on Human Behavior Bench?

The 2026-10-20 source capture contains 11 AI models.

### Does Human Behavior Bench affect BenchLM's overall score?

Not directly. Human Behavior Bench is still displayed on BenchLM for reference, but it is excluded from the weighted scoring formula.