Bulk download street-level and geospatial imagery by area.
Search a place or coordinates, choose categories, then export the Mapillary assets you need—street, intersections, roads and POIs—in one click.
Built to handle serious workloads
From pilot projects to million‑image datasets
The platform is designed to support long‑running, large‑scale Mapillary download jobs while keeping reliability and cost under control.
High task completion rate
Millions of images per job
Project‑level task history
Designed to fit into your existing geospatial stack
Connect Mapillary data with your GIS, databases and analysis pipelines
Use the platform as a reliable middle layer between Mapillary and your internal workflows, without reinventing download scripts for every project.
Mapillary API adapter
Centralised integration with Mapillary APIs, including authentication, rate limiting and error handling managed by the platform.
GIS & databases
Export clean GeoJSON/CSV metadata ready for ingestion into GIS tools and spatial databases such as QGIS, PostGIS and BigQuery.
Object storage
Store zipped imagery bundles and logs in your preferred object storage, keeping download delivery fast and auditable.
Data pipelines
Wire download jobs into existing ETL and model training pipelines using structured outputs and task webhooks or APIs.
Annotation platforms
Feed curated imagery and metadata directly into labelling tools to bootstrap map feature and computer vision projects.
Reports & dashboards
Connect usage and task statistics to BI tools so product and ops teams can monitor adoption and costs.
Core capabilities for Mapillary bulk downloads
Everything you need to turn Mapillary into a reliable internal data source
Configure once, reuse across projects. The platform abstracts Mapillary APIs, large‑scale downloads and delivery so your team can focus on analysis instead of plumbing.
Core capabilities for Mapillary bulk downloads
Configure once, reuse across projects. The platform abstracts Mapillary APIs, large‑scale downloads and delivery so your team can focus on analysis instead of plumbing.
Built for your whole geospatial team
Different roles, one shared source of Mapillary imagery
Whether you run mapping operations or train computer vision models, the platform provides a shared, self‑serve way to request and manage Mapillary data.
Built for your whole geospatial team
Whether you run mapping operations or train computer vision models, the platform provides a shared, self‑serve way to request and manage Mapillary data.
- GIS engineers: quickly assemble local datasets for map updating without maintaining brittle scripts.
- Map product managers: configure and track download tasks from a simple web UI, no engineering help required.
- Data scientists & ML engineers: download filtered, well‑structured imagery for training and evaluation datasets.
- Platform admins: monitor tenant usage, enforce quotas and protect Mapillary API limits from abuse.
Reliability and control by design
Handle large, long‑running jobs with confidence
Under the hood, the platform behaves like a purpose‑built job runner for Mapillary downloads, optimised for safety, observability and cost control.
Smart task queue & throttling
Global and per‑tenant rate limits plus queued execution protect Mapillary APIs while keeping throughput predictable.
Resumable, long‑running jobs
Jobs can pause, resume and survive transient failures, allowing very large downloads to complete without starting from scratch.
Detailed logs & partial completion
Per‑task logs, error lists and a clear "completed" / "partially completed" status help teams trust and audit every dataset.
Usage stats & quota enforcement
Track tasks, downloaded images and data volume per tenant to enforce fair‑use limits and control storage and bandwidth costs.
Scales with your workloads
Horizontally scalable workers, internal task sharding and streaming downloads keep performance stable as projects grow.
Ready for more data sources
An abstraction layer around Mapillary prepares the architecture to plug in other street‑level or remote sensing providers later.
From Mapillary to your workspace in a few clicks
Create repeatable download jobs that plug into your mapping, analytics and data science workflows without custom engineering for each project.
Simple plans that scale with your imagery needs
Start with a generous free tier, then upgrade when your projects require higher quotas, more concurrency or API access.
Free
Evaluate the platform with small areas and limited image volumes
- Up to 3 active projects
- Generous monthly image quota for evaluation
- Single user workspace
- Standard task queue and retries
- Lower per‑task image caps
- Shorter artefact retention
- No API access
Lifetime
One-time purchase for permanent access to the platform
- Custom quotas and concurrency tuned to your workloads
- Dedicated environment and SSO options
- Advanced governance, audit logs and usage dashboards
- Priority SLAs and dedicated support
- Custom integrations and onboarding
- Flexible data retention policies aligned with your compliance requirements
- Solution design support to embed the platform into your mapping and ML workflows
FAQ
Frequently asked questions about the platform
Start your first Mapillary download project
Create a project, draw an area on the map and have your first dataset ready in hours instead of days of scripting.
Testimonials
How different teams use the Mapillary Bulk Downloader
Li Wei
Senior GIS EngineerWe used to maintain several fragile Python scripts for Mapillary downloads. Moving to the platform means I configure an area once and let it handle API limits, retries and packaging. Now dataset prep is measured in minutes instead of days.
Sarah Müller
Map Product ManagerFor each new city project I used to file tickets and wait for engineering to run custom scripts. Now I can create and monitor download jobs myself and share progress with stakeholders in a single dashboard.
Diego Santos
Data Scientist, MobilityThe ability to filter by time, orientation and sampling strategy gives us focused training sets without over‑downloading. Clean GeoJSON and CSV outputs plug straight into our feature extraction and model pipelines.
Anna Kowalska
Platform AdminBefore the platform it was hard to know who was calling Mapillary and how much data we were pulling. Quotas, per‑tenant stats and a central queue give us cost control and protect us from accidentally hitting API limits.
Tomoko Sato
Urban Planning ConsultantFor corridor studies and roadside asset inventories we can quickly spin up repeatable projects for different cities. Having consistent folder structures and metadata means our GIS team can reuse the same analysis templates everywhere.
Marcus Brown
Infrastructure Lead at a mapping startupRunning large Mapillary downloads used to be a risky overnight batch job. With resumable tasks, logs and partial‑completion states, our ops team finally trusts that long‑running jobs will either finish or fail predictably.