AI Power Grid Workers

Put your local models to work. Earn AIPG.

Keep your existing setup. Connect a worker, choose when it runs, and earn AIPG for completed, accepted jobs.

Work availability and earnings vary. No signup bonus or guaranteed rate.

LM Studio, Ollama, vLLM, SGLang, LMDeploy, or another OpenAI-compatible server. The endpoint matters, not the engine.

Anthropic / Messages endpointsThe worker also serves Anthropic-format requests when the backend passes its protocol probe. This release still requires a working OpenAI chat endpoint to register; Anthropic-only endpoints are not supported yet.
2. Choose your operating system

Ubuntu 22.04+ x86_64. Select the machine that will run the worker.

3. Download your worker

Download Linux installer

v0.3.9 · Release checksums and artifact identity checked.

Release details & checksums
SHA256SUMSSPDX SBOMRelease notesDownload Linux binary directly

Get connected with OpenAI-compatible endpoint

  1. Start your inference server. Its chat endpoint and exact model must pass the worker's local checks. OpenAI-compatible endpoint setup guide
  2. Run the downloaded installer, then start the worker:
    cd ~/Downloads
chmod +x install-worker.sh
./install-worker.sh
~/.local/bin/grid-inference-worker --verify-runtime
~/.local/bin/grid-inference-worker
    The local setup wizard opens at http://localhost:7861.
  3. Select your backend and the exact model it serves in the wizard. Follow its Console approval step. Do not expose your local inference server to the internet.
  4. Wait for Online and the connection check to pass. Add your payout wallet in Console settings to receive AIPG.

Your model files stay local. Control your schedule, concurrency and pause from the worker.

The worker never needs a wallet private key. Enter credentials only in the local setup wizard, never in public posts or shell commands. Workers process plaintext prompts and outputs.

Need setup help?
Not running a model yet? Plan your setup

Operator planner

Find the useful path for your machine

Choose the workload you want to offer and adjust the example specs to match your machine. Nothing is submitted. Your local backend must still pass a generation test.

Operating system

Likely starting point

Start with the text worker and your existing backend

Start Ollama or an OpenAI-compatible local server, load a model that fits your hardware, then select and test it in the worker. VRAM alone cannot determine model fit: quantization and context length also matter.

A verified worker download is available for your selected platform. Check current network needs separately; they are not recommendations for what fits your GPU.

View worker downloads

Network-priority text route

gpt-oss-120b

1 serving worker and 2,385 jobs producing 1,216,939.63 accepted den in 30 days. Priority uses accepted den and missing replicas, not raw request count or hardware compatibility. Advertise it only when your backend genuinely serves that model.

Entered accelerator
NVIDIA
Expected text speed
Not entered
See current capacity needs and workload history

Live network opportunity

Capacity and actual 30-day work

Jobs per worker is a rough workload-share signal, not a payout forecast. Throughput is observed across the existing Grid and is not a benchmark for your GPU.

Smollm-135m

text

Single-worker risk
Workers
1
Jobs, 30d
77,438
Jobs / worker
77438.0
Observed
396.6 tok/s

deepseek-v4-flash-nvfp4

text

Single-worker risk
Workers
1
Jobs, 30d
3,008
Jobs / worker
3008.0
Observed
84.6 tok/s

gpt-oss-120b

text

Single-worker risk
Workers
1
Jobs, 30d
2,385
Jobs / worker
2385.0
Observed
53.0 tok/s

qwen3-27b

text

Single-worker risk
Workers
1
Jobs, 30d
757
Jobs / worker
757.0
Observed
75.3 tok/s

qwen38-flash-next-125b-nvfp4

text

Single-worker risk
Workers
1
Jobs, 30d
592
Jobs / worker
592.0
Observed
6.6 tok/s

FLUX.2 Klein 4B FP8

image

Single-worker risk
Workers
1
Jobs, 30d
109
Jobs / worker
109.0
Observed
3.7s average

gpt-oss-20b

text

Single-worker risk
Workers
1
Jobs, 30d
103
Jobs / worker
103.0
Observed
128.3 tok/s

LTX-2.3

video

Single-worker risk
Workers
1
Jobs, 30d
48
Jobs / worker
48.0
Observed
65.9s average

ace-step-v1.5-xl-turbo

audio

Single-worker risk
Workers
1
Jobs, 30d
28
Jobs / worker
28.0
Observed
13.1s average

z-image-turbo

image

Single-worker risk
Workers
1
Jobs, 30d
26
Jobs / worker
26.0
Observed
17.5s average

Krea 2 Turbo

image

Single-worker risk
Workers
1
Jobs, 30d
23
Jobs / worker
23.0
Observed
8.3s average

LTX Director 2.0

video

Single-worker risk
Workers
1
Jobs, 30d
0
Jobs / worker
0.0
Observed
No recent timing sample

LTX-2.3 Audio

video

Single-worker risk
Workers
1
Jobs, 30d
0
Jobs / worker
0.0
Observed
No recent timing sample
Compatibility, controls & privacy

Bring your own runtime

Keep the AI stack you already run.

The worker runs beside your inference service. It does not replace your runtime, upload your model files, or silently advertise every model it discovers.

Compatibility

What can connect today

Backend setup guides
OllamaOpen

Text · Detected automatically

vLLM / LMDeployOpen

Text · OpenAI-compatible endpoint

SGLangOpen

Text · OpenAI-compatible endpoint

LM StudioOpen

Text · Detected or entered locally

KoboldCppOpen

Text · Detected or entered locally

Other OpenAI-compatible APIsAdvanced

Text · Operator-entered endpoint

ComfyUIManual setup

Image / video · Existing ComfyUI bridge; supported models and workflows

ACE-StepQualification

Audio · Reviewed direct-runtime profile

Open describes protocol compatibility, not every model or hardware configuration. Check your platform above. ComfyUI uses the existing bridge and supported workflows; the managed installer is separate. Qualification requires a reviewed Grid profile and canary before that capability may be advertised.

You choose capacity

Text operators control the advertised model, response limits, concurrency, and operating schedule. Pause without uninstalling.

Your model stays yours

Model weights stay in your runtime. The worker needs a scoped Grid credential, never a wallet private key.

Workloads are not private yet

Community workers process plaintext prompts and outputs. Do not send secrets or regulated data until confidential execution is independently verified and available.

Media qualification progress

Needed right now

Help qualify the media worker.

The first managed media profile stays closed until accepted evidence covers every required GPU class. Each evidence set runs three local canaries and cannot enroll a worker or earn rewards.

Submit qualification evidenceView cohort status
Evidence needed

Minimum

12 GB to under 20 GB VRAM

0/1 accepted evidence sets

Evidence needed

Midrange

20 GB to under 80 GB VRAM

0/1 accepted evidence sets

Evidence needed

Datacenter

80 GB or more VRAM

0/1 accepted evidence sets

Profile ace-step-v1.5-xl-turbo v0.2.4 · status updated Sep 4, 2026 UTC

Payout history & check my worker

Public operator proof

Check the rail before you commit a GPU

Verify that workers are visible and payouts are settling. The Grid does not turn hardware specs or historical demand into guaranteed earnings.

Payout scenario

Public payout evidence is unavailable, so no estimate is shown.

Verify payouts on Base

Live worker check

Is your worker visible to the Grid?

Paste the exact worker name or worker ID printed after connection. The check reads the public online registry and sends no hardware details.

Start the worker first, then use the identity from its local dashboard or connection log.

Open public network status
How local checks and Grid connection work

Local decision

The browser does not decide what your GPU can run.

Private hardware check

The manager detects GPU, VRAM, driver, RAM, disk, and architecture locally. The Grid receives a capability tier and measured performance, not your complete hardware inventory.

Fail-closed profiles

Exact source, dependencies, model files, and recipes are verified before installation and again before serving. A capability stays unavailable until its local canary passes.

Operator flow

From your backend to Grid work

  1. 01

    Connect locally

    Start the runtime you already use, or install one separately. The worker detects its endpoint and available models locally.

  2. 02

    Choose and test

    Select what to advertise, set capacity limits, and run a real local generation before the worker connects to the Grid.

  3. 03

    Verify and serve

    Confirm the worker appears online, then serve compatible jobs. Completed accepted work contributes to the current payout split.