🚀 ZERO-GRAVITY DYNAMO

Break Free from Google Antigravity • Sovereign Google AI Account Reservoir for VS Code & Cursor
Pool Google AI Free, Pro & Ultra Plans • Zero-429 Auto-Failover • Gemini-Centric Frontier AI Reservoir • Live Quota Tachometers • Token Odometers • Local OpenAI Proxy • Native vscode.lm Provider

Overview • Architecture • Capabilities • Comparison Matrix • Quick Start • Proxy Setup • Settings

🚀 Overview: Break Free from Google Antigravity

ZERO-GRAVITY DYNAMO is designed specifically for software developers who do not want to use Google Antigravity and refuse to be chained to Google's closed ecosystem.

If you prefer coding in industry-standard IDEs like Visual Studio Code, Cursor, Windsurf, or VSCodium, you no longer have to sacrifice access to Google's best frontier AI models or suffer through a locked-down proprietary editor. Antigravity can go the way of the dodo bird—ZERO-GRAVITY DYNAMO liberates Google's AI models so you can use them directly in the IDE you already love.

🏊 The Primary Function: Multi-Account Google AI Pooling

Beyond setting you free from Antigravity, the primary function of ZERO-GRAVITY DYNAMO is to pool multiple Google AI accounts into a single, high-capacity, zero-rate-limit AI reservoir.

Whether you are using:

• Google AI Free Plans (standard free-tier developer accounts),
• Google AI Pro Plans (such as Google One AI Premium subscriptions),
• Google AI Ultra Plans (enterprise & elite high-throughput tiers), or
• Any combination of Google AI subscriptions...

ZERO-GRAVITY DYNAMO pools all your Google accounts together. Instead of crashing, freezing, or stalling with HTTP 429: RESOURCE_EXHAUSTED when an individual account exhausts its hourly or daily quota, DYNAMO transparently rotates across your pooled account tanks in real time—without dropping your active generation stream, editor cursor, or IDE connection.

♊ Centric to Google Gemini & Google AI Subscription Models

ZERO-GRAVITY DYNAMO is architected primarily around Google's frontier Gemini models:

• Gemini 2.5 Pro (flagship deep reasoning and software architecture),
• Gemini 2.5 Flash (ultra-low latency high-throughput generation),
• Gemini 3.8 Flash High (peak coding velocity & dense token streaming),
• Gemini Experimental & Reasoning Variants, and
• Any other models encapsulated by Google AI plans as Google's offerings evolve over time.

All pooled accounts feed into a unified local proxy (http://127.0.0.1:4040/v1) and a native vscode.lm provider, delivering unmetered, zero-rate-limit AI intelligence directly to VS Code Copilot, Cursor, Cline, Continue, Aider, and CLI agents.

⚡ Architecture

       ┌─────────────────────────────────────────────────────────────────┐
       │     VS Code / Cursor / Windsurf / VSCodium / Cline / Aider      │
       │       (Break Free from Google Antigravity's Closed Garden)      │
       └────────────────────────────────┬────────────────────────────────┘
                                        │
                                        ▼
       ┌─────────────────────────────────────────────────────────────────┐
       │               ZERO-GRAVITY DYNAMO POOLING CORE                  │
       │          The Sovereign Multi-Account Google AI Reservoir        │
       │                                                                 │
       │  ┌─────────────────────────────┬─────────────────────────────┐  │
       │  │ Multi-Account Quota Pool    │ Gemini-Centric Stratum      │  │
       │  │ (Zero-429 Auto-Failover)    │ (2.5 Pro, Flash, Ultra, OSS)│  │
       │  ├─────────────────────────────┼─────────────────────────────┤  │
       │  │ Zero-Cost Prompt Cache      │ Universal Smart Aliases     │  │
       │  │ (<1ms Instant Replay)       │ (dynamo/fast, reason, code) │  │
       │  ├─────────────────────────────┼─────────────────────────────┤  │
       │  │ Live Velocity Speedometer   │ Token Odometers (Trip/Life) │  │
       │  │ & Real-Time Tok/s Gauge     │ & Commercial Dollar ROI     │  │
       │  └─────────────────────────────┴─────────────────────────────┘  │
       └────────────────────────────────┬────────────────────────────────┘
                                        │
                       ┌─────────────────┼─────────────────┐
                       ▼                 ▼                 ▼
                ┌──────────────┐  ┌──────────────┐  ┌──────────────┐
                │ Google AI [#1](https://github.com/dynamo/zero-gravity-dynamo/issues/1) │  │ Google AI [#2](https://github.com/dynamo/zero-gravity-dynamo/issues/2) │  │ Google AI [#3](https://github.com/dynamo/zero-gravity-dynamo/issues/3) │
                │  (Free Plan) │  │  (AI Pro)    │  │  (AI Ultra)  │
                └──────────────┘  └──────────────┘  └──────────────┘

🎯 Key Capabilities

1. 🛡️ Sovereign Google AI Multi-Account Pool & Zero-429 Auto-Failover

• Pool Any Google AI Plan: Seamlessly aggregate accounts across Google AI Free, Google AI Pro, and Google AI Ultra tiers. Connect 2, 5, or 100+ accounts into a single unstoppable quota reservoir.
• Centric to Google Gemini: Built and tuned specifically for Gemini 2.5 Pro, Gemini 2.5 Flash, Gemini 3.8 Flash High, and any other models bundled into Google AI plans.
• Break Free from Google Antigravity: Leave proprietary locked-down IDEs behind. Run Google's frontier AI directly inside VS Code, Cursor, Windsurf, or VSCodium.
• Predictive Pre-Rotation: Continuously balances requests across the freshest pooled tanks before rate-limit thresholds trigger.
• Stratified Quota Isolation: Quotas are isolated by capability stratum (Gemini Pro, Gemini Flash, Claude, GPT-OSS, Experimental). If high-cost reasoning models hit a cooldown, your fast and standard models stay 100% active and unblocked.
• Cross-Stratum Cascade: Optional automated failover smoothly cascades exhausted queries to high-throughput backup models rather than interrupting developer workflows.
• Background Health Sweeper: Proactively refreshes OAuth access tokens 10 minutes prior to expiration and automatically unblocks tanks when quotas reset.
• Multi-Account Ingestion: Instant onboarding via Google OAuth 2.0 PKCE, direct token entry, or auto-import from local credential paths.

2. 🎛️ Live Aviation Cockpit HUD & Telemetry

• Dual-Pod Master Instrument Cluster:
  • Pod 1 (Reservoir Gauge): Cylindrical plasma fuel manifold tracking total collective reserve alongside a precision concentric dual-arc fuel tachometer.
  • Pod 2 (Velocity Tachometer): High-resolution analog/digital speedometer measuring real-time generation speed (Tok/s), moving average, peak speed record, and rolling velocity sparkline.
• Digital Token Odometers: Mechanical-style Trip and Cumulative Lifetime token counters with streaming activity indicator LEDs and one-click reset controls.
• Commercial ROI Savings Tracker: Real-time dollar calculator estimating commercial API savings against market list rates.
• Responsive Side-Wall Snapping: Designed specifically for the VS Code Activity Bar sidebar. When you drag the sidebar narrower (from 450px+ wide down to 200px thin), the interface dynamically snaps from a multi-column dashboard into a sleek single-column vertical instrument stack without clipping or horizontal scrollbars!

3. 🔌 Built-In Local OpenAI-Compatible Proxy Server

• Runs a high-performance local proxy on http://127.0.0.1:4040/v1.
• Provides 100% drop-in compatibility for Cursor, Cline, Continue, Aider, OpenDevin, LangChain, Python scripts, and standard curl.
• All requests routed through the proxy automatically inherit Zero-429 multi-tank rotation, prompt caching, and live telemetry tracking.

4. 🧠 Zero-Cost Local Prompt Cache

• Sub-Millisecond (<1ms) SSE Replays: Identical prompts return immediately from local memory without consuming network bandwidth or upstream quota.
• Deterministic SHA-256 Fingerprinting: Normalizes system prompts, messages, role configurations, temperatures, and stop sequences.
• LRU Memory Management: Keeps high-frequency developer workflows responsive while automatically discarding stale queries.

5. 🔀 Universal Smart Tier Aliases Switchboard

• Decouple client tool configurations from rigid upstream model IDs:
  • dynamo/reason: Deep logic and complex architecture.
  • dynamo/fast: High-speed generation and code completions.
  • dynamo/code: Precision programming and refactoring workhorse.
  • dynamo/free: High-throughput unmetered tier.
• 1-Click Live Re-Routing: Switch underlying models for any alias instantly in the Cockpit HUD without editing JSON configs or restarting servers.

6. 🤖 Native VS Code Language Model (vscode.lm) & Chat Participant

• vscode.lm Provider: Seamlessly exposes pooled models to GitHub Copilot and VS Code extensions.
• Brand Identity Icon Font ($(dynamo-infinity)): High-resolution vector emblem integrated into the model picker, dropdown list, and status bar.
• @dynamo Chat Participant:
  • @dynamo /status: Display active pool fuel levels, quota strata, and velocity metrics.
  • @dynamo /rotate: Manually advance to the next account tank.
  • @dynamo /reset: Clear active cooldown blocks across all tanks.
  • @dynamo /reset-odometer: Reset trip odometers and velocity peak records.
  • @dynamo /diagnose: Run real-time diagnostic checks across all connected accounts.
  • @dynamo /clear-cache: Flush the zero-cost local prompt cache.

7. 🩺 Fleet Autopilot & Reservoir Doctor

• Continuous Discovery & Health Sweep: Proactively probes Google AI Pro endpoints to verify account vitality.
• One-Click Diagnostic Doctor: Validates OAuth token scopes, token expiration, and API reachability.
• Lossless Backup & Restore: One-click encrypted JSON backup and restoration for seamless multi-machine workstation sync.

8. 🔐 Aegis Node-Locked Cryptographic Licensing

• Cryptographic Ed25519 Signatures: Offline digital signature verification against the Dynamo Master Public Key.
• Hardware-Locked Machine Identity (HWID): Multi-factor machine fingerprint bound to CPU, Motherboard, and Disk UUIDs.
• Hardware-Bound AES-256-GCM Vault: Encrypted local licensing vault (~/.gemini/dynamo_license.vault) protected against piracy and cross-machine copying.
• 7-Day Grace Protection: Uninterrupted offline capability ensures you never get locked out during network outages.

📊 Commercial Edition & Feature Comparison Matrix

🛠️ Quick Start

1. Open the Telemetry Cockpit HUD by clicking the ZERO-GRAVITY DYNAMO icon in the Activity Bar or pressing Ctrl+Alt+D C.
2. Click + Add Accounts to connect your Google AI accounts (Free, Pro, or Ultra plans) via OAuth 2.0 PKCE. Adding 2 or more accounts activates infinite automatic failover.
3. Select your active default model from the Cockpit dropdown (e.g., Gemini 3.8 Flash High, Gemini 2.5 Pro, or Claude 3.7 Sonnet).
4. Type a test prompt into the Prompt Test Console and click ⚡ Run to watch the tachometer needles and odometers spring to life!

🔌 OpenAI-Compatible Proxy Setup

ZERO-GRAVITY DYNAMO serves an OpenAI-compatible HTTP proxy on http://127.0.0.1:4040/v1. Configure your favorite tools in seconds:

Cursor / Cline / Continue (config.json)

{
  "models": [
    {
      "title": "ZERO-GRAVITY DYNAMO (Gemini 2.5 Pro)",
      "provider": "openai",
      "model": "gemini-2.5-pro",
      "apiBase": "http://127.0.0.1:4040/v1",
      "apiKey": "sk-dynamo"
    },
    {
      "title": "ZERO-GRAVITY DYNAMO (Smart Reason)",
      "provider": "openai",
      "model": "dynamo/reason",
      "apiBase": "http://127.0.0.1:4040/v1",
      "apiKey": "sk-dynamo"
    }
  ]
}

Aider

aider --openai-api-base http://127.0.0.1:4040/v1 --openai-api-key sk-dynamo --model dynamo/code

cURL Test

curl http://127.0.0.1:4040/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer sk-dynamo" \
  -d '{
    "model": "dynamo/fast",
    "messages": [{"role": "user", "content": "Explain quantum entanglement in 2 sentences."}]
  }'

⌨️ Keyboard Shortcuts

⚙️ Configuration Settings

📜 License & Compliance

Distributed under the MIT License. Engineered for high reliability, sovereign multi-key AI pooling, and unmetered developer velocity.