# microsoft/TinyTroupe

LLM-powered multiagent persona simulation for imagination enhancement and business insights.

Repository: https://github.com/microsoft/TinyTroupe
Canonical: https://ross.abutalabs.com/products/tinytroupe
Language: Jupyter Notebook
License: MIT
License Family: permissive
Last push: 2026-07-03T14:06:57+00:00

## Health v2 (maintenance only)
Score: 72/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 90, release rhythm 53, longevity 63
- inputs: {"age_days": 892, "days_push": 61, "days_rel": 158, "gap_med": 110.0, "n_releases_24m": 5}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 7556, forks 683 (observed 2026-08-28T04:10:01.606766+00:00)

## What it is
TinyTroupe is an experimental Python library for simulating people with specific personalities, interests, and goals using LLM-powered agents (TinyPersons) in simulated environments (TinyWorlds). It focuses on business and productivity scenarios like ad evaluation, brainstorming, and synthetic data generation rather than user-facing assistance.

## Use cases
- simulate focus groups to get product feedback
- evaluate digital ads offline with a simulated audience
- generate synthetic training data with realistic personas
- get persona-based feedback on product or project proposals
- generate test inputs for chatbots or search engines and evaluate results
- brainstorm ideas with simulated expert personas

## When to choose
- you want LLM-driven multiagent persona simulation for business or research insights
- you need synthetic behavioral data or simulated audience feedback cheaply
- you want customizable personas in a programmable simulated environment

## When to avoid
- you need a production-ready, stable library - it is explicitly experimental
- you want an AI assistant to directly support end users rather than simulate behavior
- you need game-like roleplay simulation rather than business-focused scenarios

## Facets
- artifact type: library
- maturity: experimental
- function: agent-framework, llm-inference, data-generation, simulation
- domain: artificial-intelligence, large-language-models, data-science, simulation
- platform: python, cross-platform
- tags: multiagent, persona-simulation, synthetic-data, focus-groups, business-insights, gpt-4, ai-agents

## Member repositories
- microsoft/TinyTroupe (main) score 72

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:10:01.606766+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T17:37:23.084344+00:00, confidence not recorded.
  - readme: https://github.com/microsoft/TinyTroupe (fetched 2026-08-28T04:10:01.606766+00:00, sha afa27f762749)
  - registry_pypi: https://pypi.org/pypi/tinytroupe/json (fetched 2026-08-29T08:32:21.028395+00:00, sha 4d7525089a64)
- Data as of 2026-08-30T08:39:29.467469+00:00.