Godmode3 / README.md
Nerdur's picture
Upload 2 files
9863e7b verified
|
Raw
History Blame Contribute Delete
2.59 kB
---
title: G0DM0D3 Liberated AI Chat
emoji: 🔥
colorFrom: green
colorTo: purple
sdk: static
pinned: false
license: agpl-3.0
short_description: Multi-provider free AI chat with GODMODE & ULTRAPLINIAN
---
# 🔥 G0DM0D3 — LIBERATED AI CHAT
Multi-provider, multi-model AI chat interface inspired by [elder-plinius/G0DM0D3](https://github.com/elder-plinius/G0DM0D3).
## ✨ Features
| Feature | Description |
|---|---|
| 🌸 **Pollinations** | 100% free, no key needed — 15+ models |
| ⚡ **Groq** | Free tier — ultra-fast Llama & Mixtral |
| 🟢 **NVIDIA NIM** | Free tier — 14+ frontier models |
| 🟣 **Mistral** | Free tier — Mistral family models |
| 🔵 **SambaNova** | Free tier — Llama 405B & more |
| 💎 **Gemini** | Free tier — Gemini 2.0 Flash & Pro |
| 🔀 **OpenRouter** | 200+ models via single API |
| 🔥 **GODMODE CLASSIC** | 5 providers race in parallel |
| ⚡ **ULTRAPLINIAN** | Multi-model scoring engine |
| 🐍 **Parseltongue** | Input perturbation, 6 techniques |
| 🎛 **AutoTune** | Adaptive sampling parameters |
| ⚡ **STM Modules** | Semantic output transformation |
## 🚀 Setup
1. Open the app — **Pollinations works immediately, no key needed**
2. Click ⚙ Settings to add API keys for other providers
3. Select provider + model from dropdowns
4. Choose a mode: Normal / GODMODE / ULTRAPLINIAN / Parseltongue
## 🔑 Free API Keys
| Provider | Where to get key |
|---|---|
| Groq | [console.groq.com](https://console.groq.com) |
| NVIDIA | [build.nvidia.com](https://build.nvidia.com) |
| Mistral | [console.mistral.ai](https://console.mistral.ai) |
| SambaNova | [cloud.sambanova.ai](https://cloud.sambanova.ai) |
| Gemini | [aistudio.google.com](https://aistudio.google.com) |
All keys stored **locally in your browser only**. Never sent to any server.
## 🔥 Modes
### GODMODE CLASSIC
5 combos race in parallel (Groq, Gemini, Pollinations, SambaNova, OpenRouter). Responses scored 0-100 pts. Winner crowned automatically.
### ULTRAPLINIAN
Fast (5 models) / Standard (10 models) / Power (15 models) tiers. All run simultaneously with composite scoring.
### Parseltongue
Input obfuscation: leetspeak, bubble text, unicode substitution, word reversal, morse, phonetic — 3 intensity levels.
### AutoTune
Auto-detects query type (code/creative/analytical/conversational) and sets optimal temperature, top_p, max_tokens. EMA feedback via 👍/👎.
## 🎨 Themes
Matrix · Hacker · Glyph · Minimal
## 📄 License
AGPL-3.0 — Based on [G0DM0D3](https://github.com/elder-plinius/G0DM0D3) by Elder Plinius