Stay free if you only need limited vectorization credits that reset monthly and ai line autocomplete and polygon tracing. Upgrade if you need custom credit volumes for large teams or bulk digitization projects and dedicated support and onboarding assistance. Most solo builders can start free.
Why it matters: Requires internet connectivity because inference runs on Bunting Labs' cloud servers—no offline or air-gapped mode
Available from: Pro
Why it matters: Sends raster data to a third-party server, which may not be acceptable for classified, defense, or legally sensitive cadastral workflows
Available from: Pro
Why it matters: Only integrates with QGIS; no ArcGIS Pro, MapInfo, or standalone CLI version is documented
Available from: Pro
Why it matters: Accuracy, by the company's own admission, has not yet exceeded human performance, so complex or noisy maps still require cleanup
Available from: Pro
Why it matters: Pricing tiers and exact feature gating are not published on the blog post—users must sign up to see paid plan details
Available from: Pro
Open QGIS, go to the Plugins menu, and search the official QGIS plugin repository for "Bunting Labs AI Vectorizer." Install it and a new vectorize icon will appear in your plugin toolbar. Click the icon to start the onboarding flow, which prompts you to sign up with your email and confirm your account. If you already registered on the Bunting Labs website, you instead paste a Secret Key from your dashboard to authenticate.
No. The machine learning model runs remotely on Bunting Labs' web servers, not on your local machine. If your computer can run QGIS itself, it can run the AI Vectorizer plugin with no additional hardware requirements. Traffic between the plugin and Bunting Labs' servers is encrypted, but because inference is cloud-based you do need a reliable internet connection while digitizing.
It handles both. For lines, you click two points to seed a direction and the AI completes as much of the line as it is confident about. For polygons, you activate the plugin on a polygon layer and it traces the feature's border and fills the interior automatically. In either case, you right-click to commit the finished geometry to your vector layer, and you can use the shift key to correct the trace if the AI veers onto the wrong feature.
According to Bunting Labs, the model on average keeps a traced line within the same pixel that a GIS professional would choose manually. The company explicitly states it has not yet exceeded human performance, so it should be viewed as an autocomplete that accelerates a skilled human rather than a fully autonomous digitizer. Against older non-ML QGIS plugins like AutoTrace (2015) and ContourTrace (2011), Bunting Labs reports roughly a 10x throughput improvement.
New users start on a free tier that provides a limited number of vectorization credits each month at no cost. The Pro tier is priced at $29 per month and unlocks a higher monthly credit allocation, increased digitization throughput, and fine-grained control over how maps are stored on Bunting Labs' servers. Enterprise pricing is custom-quoted for large teams needing bulk credit volumes, dedicated support, and custom SLAs. Users can confirm current rates at buntinglabs.com/pricing. Based on our analysis of 870+ AI tools, credit-gated freemium is a common pattern in specialized GeoAI plugins.
Start with the free plan — upgrade when you need more.
Get Started Free →Still not sure? Read our full verdict →
Last verified March 2026