Skip to content
Provider Worker Releases

Download GPU Contribution Bundles

Version 0.2.0 . Download your platform bundle, benchmark hardware, start worker mode, and begin earning ANM.

Provider Overview Need Installation Help

windows

Windows Worker Bundle

Includes launcher scripts, config template, logs folder, and executable placeholder.

Version
0.2.0
File
aicf-provider-worker-0.2.0-windows-x64.zip
Size
N/A
SHA256
pending
Download Release Notes

linux

Linux Worker Bundle

Includes benchmark/start scripts and systemd unit template.

Version
0.2.0
File
aicf-provider-worker-0.2.0-linux-x64.tar.gz
Size
N/A
SHA256
pending
Download Release Notes

python

Python Source Bundle

Source-first option for custom runtime instrumentation and local modifications.

Version
0.2.0
File
aicf-provider-worker-0.2.0-python.tar.gz
Size
N/A
SHA256
pending
Download Release Notes

Install · Windows Worker Bundle

  1. Extract ZIP and open PowerShell in the bundle folder.
  2. Run benchmark first: `./benchmark-worker.bat`.
  3. Generate config: `./aicf-provider-worker.exe init-config --config provider.config.json`.
  4. Start worker: `./start-worker.bat`.
benchmark-worker.bat
aicf-provider-worker.exe init-config --config provider.config.json
start-worker.bat

Install · Linux Worker Bundle

  1. Extract archive: `tar -xzf aicf-provider-worker-<version>-linux-x64.tar.gz`.
  2. Make launchers executable: `chmod +x *.sh`.
  3. Run benchmark: `./benchmark-worker.sh`.
  4. Start daemon: `./start-worker.sh`.
chmod +x benchmark-worker.sh start-worker.sh
./benchmark-worker.sh
./start-worker.sh

Install · Python Source Bundle

  1. Extract Python bundle and create virtual environment.
  2. Install requirements: `pip install -r requirements.txt`.
  3. Benchmark hardware: `python worker.py benchmark --config provider.config.json`.
  4. Start worker: `python worker.py start --config provider.config.json`.
python -m venv .venv
. .venv/bin/activate
pip install -r requirements.txt
python worker.py benchmark --config provider.config.json
python worker.py start --config provider.config.json

Supported GPUs and Minimum Requirements

  • NVIDIA driver 535+ recommended for production queues
  • CUDA-compatible GPUs with at least 12GB VRAM for baseline inference queues
  • 16GB system RAM minimum, 32GB recommended
  • Stable internet + open outbound HTTPS to AICF API
  • Wallet payout address on Animica mainnet

Benchmark Before Joining

Worker benchmark mode detects your hardware and emits throughput metrics used by scheduler assignment. Do not skip this step. 

aicf-provider-worker benchmark --config provider.config.json
# verifies GPU detection, throughput, memory, and health checks

Provider Earnings and Economics

Providers earn ANM from completed workloads. Better benchmark class, lower failure rate, and strong uptime increase routing priority and earnings consistency.

  • Jobs escrow ANM before compute starts
  • Receipts settle rewards and update reputation
  • Treasury subsidy can boost strategic job classes

Troubleshooting

  • If no GPU is detected, verify drivers and rerun benchmark mode.
  • If wallet binding fails, regenerate config and confirm payout address format.
  • If heartbeat fails, inspect `logs/provider-worker.log` and network ACLs.
  • If jobs are not assigned, review reputation, stake, and benchmark tier.
Provider Docs Open Support