Generate structured fictional character profiles, NPC system prompts, and image prompts using local or cloud LLMs.
Create deep TRPG and novel character profiles, system prompts, and visual art prompts with custom LLMs.
CharGen is a browser-based application designed for fiction writers, game designers, and TRPG creators who need detailed character backgrounds. You can generate psychological profiles, system prompts for conversational roleplay, and image prompts for visual art using your choice of API keys or local AI models.
CharGen crafts comprehensive character sheets containing MBTI personality traits, background stories, dark secrets, and adjustable anti-cliché levels. It also outputs copy-ready system prompts for roleplaying conversations and visual art prompts structured for image generators like Midjourney and Stable Diffusion.
This tool suits novelists, game developers, narrative designers, and tabletop roleplayers who want to accelerate character creation without starting from a blank page. It is not designed for full book drafting, manuscript editing, or non-fiction documentation.
Connect directly to OpenAI, Anthropic, Gemini, Chinese LLMs, or your local Ollama instance.
Adjust character originality across five intensity levels to move from classic archetypes to extreme subversions.
Generate Stable Diffusion and Midjourney prompts tailored for character turnarounds, expression sheets, and action poses.
Export tailored system instructions that turn any standard chat interface into your generated character.
CharGen is free and open source under the MIT license. The only operational costs are the third-party API usage fees from providers like OpenAI or Anthropic, or the local computing hardware required for Ollama.
879 open remote AI-training roles, updated November 15, 2026; most pay $20–$175 an hour. Zekai lists them free; the application goes to the employer’s platform.
These reviewed AI tools do the same job instantly — from our scored directory.
Yes, you can configure CharGen to connect to a local Ollama instance running open-source models on your own hardware.
No, the app runs as a client-side React application and saves your configuration settings inside your local browser storage.
No, it generates descriptive English prompts optimized for image generation tools like Midjourney and Stable Diffusion.
See Zekai first in Google