Answer in 30 seconds
Import documents, review source fields, and export checked results in your web workspace on Windows. Connect Ollama with the Microsoft Store companion for local model inference, or choose hosted AI and your own provider keys.
Run local Ollama document extraction and image OCR on your Windows PC. Connect the companion to your web workspace, review PDF fields and image data, then export checked results.
Keep the Windows companion and Ollama running while you use local extraction. Import, source review, corrections, and exports take place in your browser workspace.
Internet access is required for the web workspace, sign-in, and account services. Ollama model inference runs locally on your PC. Download Ollama and your chosen model separately.
A Windows download does not make every processing route local. Choose the route for the document, check your plan, and review the result against the source. A cloud provider key sends inference to that provider. Local Ollama uses the companion path, while selectable-text PDF-to-Markdown uses browser parsing without AI.
Workspace and runtime setup
PDFs and scanned documents
Extract supported fields and tables, then inspect rows beside the original pages. The model and file determine the available processing path. Images and scans require an image-capable route.
Document images
Review screenshots, receipt photos, and scans for focus, missing edges, and glare before extraction. The Windows browser does not recover data that the image does not show.
Selectable-text PDFs
Use the PDF-to-Markdown tool when you need editable text without AI. It stays in the browser and charges one PE per successfully converted page on every plan.
Choose a Windows processing route
Use your browser for document import, review, and exports. The Windows companion connects the workspace to Ollama, which you install separately. Check current entitlements on the pricing page.
- Route
- Hosted AI
- Setup
- Browser workspace
- Inference or parsing
- Hosted provider through the web service
| Route | Setup | Inference or parsing |
|---|---|---|
| Hosted AI | Browser workspace | Hosted provider through the web service |
| Own provider key | Connect a supported cloud provider | Your chosen provider through the web service |
| Local Ollama | Windows companion, Ollama, suitable model, eligible plan | Local model through the companion |
| PDF to Markdown | Selectable-text PDF in the browser | Browser parsing without AI |
Check before connecting
- Route
- Hosted AI
- Setup
- Browser workspace
- Inference or parsing
- Hosted provider through the web service
- Install DynamiteDocs by SevenOneLabs from Microsoft Store and check the listed system requirements.
- Install Ollama separately and confirm that the chosen model supports the document input you want to process.
- Sign in on an eligible plan and connect the companion through the workspace AI settings.
- Run one representative file and compare fields, row boundaries, and totals with the source before exporting.
Set up and test extraction
- 01Open your workspace
- 02Add the companion for Ollama
- 03Review and export
Open your workspace
Sign in to Dynamite Docs in your browser, then try a supported document with hosted AI or a connected provider key.
Add the companion for Ollama
Install the Windows companion from Microsoft Store. Start Ollama and a suitable model separately, then connect from AI Settings on an eligible plan.
Review and export
Correct the accepted fields and rows in your browser workspace, then choose an available export. Excel, CSV, and JSON downloads require signup; plan and format capabilities still apply.
Use the accepted output
Use the same reviewed tables for Excel, CSV, or JSON regardless of the chosen inference route. Local model performance depends on your hardware, model, and input. Test the files you actually receive rather than assuming a successful text prompt proves scan extraction quality.
Processing choices and allowances
Choose hosted AI, your own provider keys, or local Ollama on an eligible plan. Each successfully processed hosted page uses one PE. Free and Hobby BYOK use the same PE allowance; Pro and Ultra BYOK cost zero PE, with provider charges and limits still applying. Failed pages cost zero PE.
Check current plans, page limits, and capabilities and review storage and data-residency policies before processing sensitive records. Local model inference is a separate choice from browser-local file storage.
Workflow boundaries
- Get the Ollama companion from Microsoft Store and use it with the Dynamite Docs web workspace.
- Check the Windows and hardware requirements of both the companion and your chosen Ollama model before installing.
- Internet access is required for the web workspace, sign-in, and account services. Ollama model inference runs locally on your PC. Download Ollama and your chosen model separately.
- Sign in to Dynamite Docs in your browser. On an eligible plan, open AI Settings and connect the local companion to select an Ollama model.
Get Dynamite Docs for Windows
The Windows companion connects your Dynamite Docs workspace to a local Ollama model. Install Ollama and a suitable model separately, then connect through AI Settings. A Dynamite Docs account and an eligible plan are required for local processing.
Questions about this workflow
What does the Windows companion do?
The companion connects your web workspace to Ollama running on your Windows PC. Use your browser for document import, source review, corrections, and exports.
Do I need Ollama for hosted AI?
No. Ollama is needed for the local companion route. Hosted AI and connected cloud provider keys use the web service.
Is every Windows workflow offline?
No. Cloud inference uses network services. Local inference uses the companion, but account access, usage reporting, and integrations can still require a connection.