Skip to main content
Back

Pliny the Prompter's G0DM0D3 Blasts 51 AI Models at Once

G0DM0D3 is a single-file, open-source interface that races 50+ AI models in parallel with built-in red-teaming and jailbreak research tools

Pro

Repo, Full Access
  • What it is: G0DM0D3 is an open-source, single-file browser app that races 50+ AI models in parallel and scores their responses.
  • Key modes: GODMODE CLASSIC (5 models), ULTRAPLINIAN (up to 51 models with composite scoring), CONSORTIUM (hive-mind synthesis), and Standard.
  • Red-teaming toolkit: Parseltongue applies 33 input perturbation techniques (leetspeak, Unicode, Morse) to probe model safety classifiers.
  • Zero install, zero login: Entire app is one index.html file; API key and chat history stay in browser localStorage only.
  • Cost model: G0DM0D3 is free (AGPL-3.0); inference costs are metered via your own OpenRouter API key — 51 parallel calls at ULTRAPLINIAN tier 5 adds up fast.
  • Limitations: No persistent memory, no IDE/codebase integration, no voice I/O; clearing browser data wipes all history.

G0DM0D3 is a fully open-source, browser-based chat interface that does something no mainstream AI product will: it fires your prompt at up to 51 models simultaneously, scores every response, and hands you the winner. Built by Pliny the Prompter (elder-plinius on GitHub), the project has accumulated over 6,500 stars and 1,500 forks since it appeared on GitHub, and it comes packaged as a single index.html file you can run by double-clicking it.

One File to Rule Them All

The entire client application is a single index.html file. No npm install, no build step, no framework dependencies , you clone the repository and open the file directly in a browser, or serve it locally with a one-line Python command. Under the hood it is roughly 3,300 lines of TypeScript compiled into that file, with an optional Node.js/Express API server in api/ for deployments that need server-side orchestration.

There is no login system and no server-side storage. Your API key lives in localStorage, your chat history lives in localStorage, and nothing is sent to G0DM0D3's infrastructure. The trade-off is that clearing your browser data wipes everything. There is a built-in export/import function, but the backup responsibility is entirely yours.

The Model Gateway: OpenRouter

G0DM0D3 does not connect to Anthropic, OpenAI, or Google directly. It routes every model call through OpenRouter, a unified API gateway that aggregates access to 50+ models under a single API key and billing account. You register at OpenRouter, generate an API key, and paste it into G0DM0D3's settings. From that point, Claude, GPT-5, Gemini, Grok, Mistral, LLaMA, DeepSeek, Qwen, and more are all available through the same interface.

Costs are metered. Running GODMODE CLASSIC fires 5 parallel model calls. Running ULTRAPLINIAN at full scale fires up to 51. Those are real API calls at real per-token prices, aggregated on your OpenRouter account. G0DM0D3 itself is free and AGPL-3.0 licensed , the inference bill is yours.

Four Modes, Four Philosophies

G0DM0D3 operates as a modular system that pushes the limits of the post-training layer by providing tools for systematic inference-time safety evaluation. That modularity surfaces as four distinct execution modes, each with a different scope and cost profile.