Verified AI compute mesh powered by Quai. Mine QUAI when idle. Serve verified dense-model inference when scheduled.
Quai Intelligence is a decentralized inference network optimized for dense-model serving. Workers mine QUAI by default, pause for paid inference work when scheduled, and prove correctness with Terkle-backed verification. Optional privacy modes keep the architecture usable for more sensitive workloads without making confidentiality claims the system cannot support on anonymous public GPUs.
Quai Intelligence is:
- a decentralized inference network
- optimized for dense-model serving
- equipped with optional privacy modes
- backed by Terkle verification
- designed so GPUs mine QUAI when idle and switch to inference work when scheduled
Quai Intelligence is not:
- a general AI utility mesh
- a storage or memory network
- confidential compute on anonymous public GPUs
┌─────────────────────────────────────────────────────────────┐
│ Quai Intelligence │
├─────────────────────────────────────────────────────────────┤
│ │
│ Users Coordinator │
│ ───── ─────────── │
│ Pay Qi for ───────► Routes requests ───────► │
│ inference to GPU miners │
│ │
│ Miners │
│ ────── │
│ Run models, earn Qi │
│ Distributed globally │
│ │
└─────────────────────────────────────────────────────────────┘
- Verified dense-model inference via Terkle-backed attestations, challenges, and semantic checks
- Hybrid worker economics with default QUAI mining plus scheduled inference work
- Sharded model serving for larger dense models across multiple miners
- Optional privacy modes for edge-owned first/last layers
- Desktop and browser UI for network visibility and miner control
quai-intelligence is the active home of the miner product.
internal/miner/contains the Go worker, local-mode logic, coordinator integration, and inference backend orchestrationkawpowminer/is the vendored low-level mining runtime fork used by the worker- the canonical upstream for mining-runtime changes is the owned fork based on
0xalank/kawpowminer - separate prototype repos such as
quai-miner-inferenceare not the ongoing implementation target
In practice this means:
- memory management, pause/resume policy, and worker-facing performance improvements are planned and integrated from this repo
- low-level CUDA / VRAM / miner-runtime changes land in the
kawpowminerfork - high-level product work such as local mode,
vllm, coordinator connectivity, and API serving stays inquai-intelligence
| Platform | Download |
|---|---|
| macOS | Quai-Intelligence-0.1.0.dmg |
| Windows | Quai-Intelligence-0.1.0.msi |
| Linux | Quai-Intelligence-0.1.0.AppImage |
# Clone
git clone https://github.com/dominant-strategies/quai-intelligence
cd quai-intelligence
# Build Go binaries
make build
# Build desktop app
cd app
npm install
npm run tauri buildquai-intelligence/
├── app/ # Desktop application
│ ├── src/ # React UI components
│ │ ├── components/
│ │ │ ├── Dashboard.tsx # Main dashboard
│ │ │ ├── GlobeNetwork.tsx # 3D world map
│ │ │ ├── GpuMonitor.tsx # GPU performance charts
│ │ │ ├── MiningPanel.tsx # Mining statistics
│ │ │ ├── ModelsPanel.tsx # Model management
│ │ │ └── SettingsPanel.tsx # Configuration
│ │ └── hooks/
│ ├── src-tauri/ # Rust backend
│ │ ├── src/main.rs # System integration
│ │ └── binaries/ # Bundled Go executables
│ └── package.json
│
├── cmd/ # Go entrypoints
│ ├── coordinator/ # Network coordinator
│ └── miner/ # GPU miner
│
├── internal/
│ ├── coordinator/ # Routing, scheduling, verification, billing
│ ├── miner/ # Worker orchestration and local hybrid runtime
│ ├── p2p/ # Data-plane transport
│ └── sglang/ # Sharded inference runtime client
│
├── kawpowminer/ # Owned mining runtime fork vendored into this repo
│
├── docker/ # Container configs
├── scripts/ # Build & release scripts
├── Makefile # Build commands
└── go.mod
-
Coordinator — Routes inference requests to available miners
- Load balancing across GPU tiers
- Model-to-miner matching
- Payment settlement
-
Miner — Runs on GPU machines
- Orchestrates mining and inference from
internal/miner/ - Launches and controls the vendored
kawpowminerruntime - Downloads and serves AI models
- Processes inference requests
- Reports hashrate, memory, and availability
- Orchestrates mining and inference from
-
Desktop App — User interface
- GPU monitoring and charts
- Model download manager
- Wallet and earnings tracker
- Global network visualization
Large models are split across multiple GPUs:
User Request
│
▼
┌─────────────┐
│ Coordinator │
└─────────────┘
│
├──────────────┬──────────────┐
▼ ▼ ▼
┌─────────┐ ┌─────────┐ ┌─────────┐
│ Miner A │──►│ Miner B │──►│ Miner C │
│ Layers │ │ Layers │ │ Layers │
│ 0-20 │ │ 21-40 │ │ 41-60 │
└─────────┘ └─────────┘ └─────────┘
│
▼
Response
Settings are stored in ~/.quai/config.json:
{
"wallet": "0x...",
"coordinator": "coordinator.quai.network:3334",
"models": ["Qwen/Qwen2.5-7B-Instruct"],
"gpu": {
"devices": [0],
"powerLimit": 100,
"tempLimit": 83
},
"autoStart": true
}The miner supports any Quai-compatible stratum pool. Default is StratumX.
| Region | Endpoint |
|---|---|
| US | stratum+tcp://mining-us.stratumx.org:3333 |
| EU | stratum+tcp://mining-eu.stratumx.org:3333 |
| Singapore | stratum+tcp://mining-sg.stratumx.org:3333 |
| Hong Kong | stratum+tcp://mining-hk.stratumx.org:3333 |
| India | stratum+tcp://mining-in.stratumx.org:3333 |
| Iceland | stratum+tcp://mining-is.stratumx.org:3333 |
| Brazil | stratum+tcp://mining-bz.stratumx.org:3333 |
| Poland | stratum+tcp://mining-po.stratumx.org:3333 |
# Use a specific pool
./quai-miner --pool stratum+tcp://mining-eu.stratumx.org:3333 --wallet 0xYourAddress
# Or in config.json
{
"pool": "stratum+tcp://mining-us.stratumx.org:3333"
}The coordinator exposes an OpenAI-compatible API:
curl http://localhost:3333/v1/chat/completions \
-H "Content-Type: application/json" \
-d '{
"model": "Qwen/Qwen2.5-7B-Instruct",
"messages": [{"role": "user", "content": "Hello!"}]
}'| Model | Size | VRAM Required |
|---|---|---|
| Qwen 2.5 7B | 14 GB | 13 GB |
| Llama 3.3 70B | 126 GB | 126 GB (multi-GPU) |
| Mixtral 8x7B | 87 GB | 87 GB |
| DeepSeek V3 | 1.1 TB | Distributed |
# Run coordinator locally
make run-coordinator
# Run miner (connects to coordinator)
make run-miner MODEL=Qwen/Qwen2.5-7B-Instruct
# Run desktop app in dev mode
cd app && npm run dev
# Run with Tauri (native window)
cd app && npm run tauri dev# Build all platforms
make release
# Outputs:
# - dist/Quai-Intelligence-0.1.0-x86_64.dmg (macOS)
# - dist/Quai-Intelligence-0.1.0-x86_64.msi (Windows)
# - dist/Quai-Intelligence-0.1.0-x86_64.AppImage (Linux)- Fork the repo
- Create a feature branch
- Submit a pull request
See CONTRIBUTING.md for guidelines.
MIT License — see LICENSE