For Universities and Libraries

SpatialRead is a desktop application designed to help students, researchers, and faculty members quickly understand complex research papers through AI-powered explanations. Users can upload research paper PDFs, highlight confusing text, and receive contextual AI explanations that make dense academic content more accessible.

All features in SpatialRead are free to use with local models via Ollama. Integration with LM Studio and cloud AI model providers like OpenAI, Google, Perplexity, is available as a one-time purchase ($49). No subscription or recurring fees.

Who It's For

• Students: Graduate and undergraduate students working with research literature
• Researchers: Academics conducting literature reviews and staying current with their field
• Faculty: Professors and instructors who need to quickly understand papers outside their expertise
• Librarians: Information professionals helping patrons navigate complex research

Key Use Case

The primary workflow is simple: add a PDF, highlight confusing text, click "Explain," and receive AI-powered explanations in the context of the research paper. This dramatically reduces the time needed to understand complex academic content.

Accelerated Research Comprehension

SpatialRead helps users understand research papers faster by providing instant explanations of complex concepts, terminology, and methodologies. This is particularly valuable for:

• Cross-Disciplinary Research: Grasping concepts from fields outside one's expertise
• Literature Reviews: Quickly understanding multiple papers across different domains
• Time-Constrained Reading: Meeting deadlines while maintaining comprehension quality
• Non-English Speakers: Understanding research papers in languages other than English

AI-Powered Contextual Explanations

Unlike generic AI tools, SpatialRead provides explanations that are contextualized within the specific research paper being read. This ensures that:

• Explanations reference the paper's specific methodology and findings
• Technical terms are explained in the context they're used
• Complex relationships between concepts are preserved

Support for Multiple Research Workflows

SpatialRead supports various research approaches:

• Spatial Canvas: Visual organization of ideas and connections between concepts
• Highlight-to-Explain: Instant explanations of selected text
• Simplify/Explain/Expand Action buttons: Depending on the user's needs, confusing text can be simplified, explained, or expanded upon. Custom questions can also be asked.
• Multi-Model Support: Choose from 300+ AI models (OpenAI, Anthropic, Google, Perplexity, and OpenRouter). Different models have different capabilities.
• Local Processing: Unlimited local model use via Ollama and LM Studio

SpatialRead is no-sign up desktop application that researchers can download and use completely for free. No emails or user identifiable information is collected.

Local-First Processing

All document processing happens locally on the user's device:

• Documents Never Leave Device: PDFs, highlights, and AI conversationsare stored locally
• No Cloud Storage: The application does not send any user data to external servers
• Complete Privacy: Users maintain full control over their research materials
• Offline usage: Works without an internet connection (when using local models)

Bring Your Own Keys (BYOK) Model

If users want to use cloud AI models, they can use SpatialRead's Bring Your Own Keys (BYOK) approach, meaning:

• User-Controlled API Keys: Students and researchers use their own API keys from AI providers
• No Data Collection: We never have access to user documents or AI interactions
• Direct Provider Communication: AI queries go directly to the user's chosen provider

Anonymous Usage Analytics

SpatialRead collects anonymous usage analytics to improve app stability and feature development. This includes:

• Feature usage events: Events like "canvas created," "node created," and "AI response generated"
• Performance metrics: AI response times, load times, error rates, and app stability metrics
• General session data: App version, operating system, and basic device information

What we never collect:

• PDF content or document metadata
• Chat messages, annotations, notes, or research content
• Personal identifiers (names, emails)
• User research data or academic work

Analytics provider and compliance:

• Analytics are processed by PostHog, which is SOC 2 Type II certified and GDPR/CCPA compliant
• IP addresses are anonymized at collection—no IP tracking or cross-session user profiling
• Analytics are used solely for product improvement and are never sold or shared with third parties
• Data residency controls ensure compliance with institutional requirements
• Deletion requests can be honored through PostHog's built-in privacy controls

Why this matters for libraries:

Usage analytics are standard practice for educational tools (even highly privacy-focused ones use them). What matters is the distinction between tracking product health metrics (like "PDF uploaded") versus tracking user behavior or content (like "user discussed quantum physics in chat message X"). SpatialRead only tracks the former, ensuring we can improve the product without surveilling user research.

Compliance-Friendly

This architecture makes SpatialRead suitable for institutions with strict privacy requirements:

• FERPA-friendly: Designed to avoid FERPA concerns by not collecting student data or personally identifiable information
• No HIPAA concerns: No health data transmission or storage
• GDPR/CCPA compliant: Analytics provider (PostHog) meets international privacy standards
• Compatible with institutional data protection policies: Local-first architecture and transparent analytics
• Suitable for sensitive research projects: Documents and research content never leave the device (and can work offline)
• No behavioral surveillance: Analytics measure product health, not user behavior for profiling or monetization

Desktop Application

SpatialRead is a desktop application available for:

• macOS: Native application with full feature support
• Windows: Native application with full feature support

Local Model Support

For unlimited, private AI usage:

• Ollama Integration: Users can run local AI models via Ollama
• No API Costs: Completely free when using local models
• Offline Capable: Works without internet connection
• Complete Privacy: All processing happens locally on the user's device and no data is sent to external servers

AI Provider Integration (Optional)

Users can use API keys from one or more AI providers:

• OpenAI: GPT-5.1 Thinking, GPT-5.1 Instant, GPT-5 Pro, GPT-5, GPT-5 Mini, GPT-5 Nano
• Google: Gemini 3 Pro, Gemini 2.5 Pro, Gemini 2.5 Flash, Gemini 2.5 Flash Lite
• Perplexity: Sonar Pro, Sonar Reasoning Pro, Sonar Reasoning, Sonar
• Anthropic: Claude Opus 4.1, Claude 4.5
• OpenRouter: Access to 300+ models from various providers including Grok, Mistral, LLama, etc.

System Requirements

• Modern operating system (macOS 10.15+, Windows 10+, or recent Linux distribution)
• Internet connection (for API-based AI models) or Ollama/LM Studio installation (for local models)
• Sufficient disk space for application and local data storage

Technical and Usage

Q: Does SpatialRead require an internet connection?

A: No. SpatialRead works completely offline when using local AI models via Ollama or LM Studio. Internet connection is only required if users choose to use cloud AI providers (OpenAI, Anthropic, Google, Perplexity, OpenRouter) with their own API keys.

Q: What operating systems are supported?

A: SpatialRead is available as a native desktop application for macOS and Windows. Linux support may be available—please contact us for current compatibility information.

Q: How do students get started with SpatialRead?

A: Students can download SpatialRead directly from spatialread.com. No account creation or email required. They can start using it immediately with local models (Ollama) for free, or optionally purchase a lifetime license to use their own API keys for cloud AI models.

Q: Is there a cost for students to use SpatialRead?

A: All features in SpatialRead are free to use with local models via Ollama. Integration with LM Studio and cloud AI model providers (OpenAI, Google, Perplexity, etc.) is available as a one-time purchase ($49). We offer a 20% student discount for verified students.

Q: Are cloud models expensive?

A: Cloud model costs vary by usage and model choice. Let's walk through an example of what costs could look like.

The average research paper is 20 pages long. Cloud models typically compress PDFs resulting in around 250 tokens per page. This means a 20 page PDF would be around 5000 input tokens. A response is typically 200 output tokens. If you use a model like Gemini 2.5 Flash Lite, which costs $0.10 per 1M input tokens and $0.40 per 1M output tokens, that would cost $0.0005 + $0.00008 = $0.00058/chat. Assuming you ask 50 questions a day, that's $0.029/day. That adds up to just $0.87/month. If you use one of the smartest and most expensive models, like GPT 5.1 Thinking ($1.25/$10 per 1M tokens), you could see your costs rise to $12.50/month.

Most services charge a flat $10-30/month fee and generally provide access to cheap models but the advantage of Bring Your Own Keys (BYOK) is that you only pay for what you use.

Q: Can SpatialRead help non-English speakers understand research papers?

A: Yes. SpatialRead can help users understand research papers in languages other than English. Many AI models supported by SpatialRead (including OpenAI, Google Gemini, and Anthropic Claude models) have multilingual capabilities and can provide explanations in various languages. Users can ask questions and receive explanations in their preferred language, making complex research papers more accessible regardless of the paper's original language or the user's native language. This is particularly valuable for international students and researchers working with papers outside their primary language.

AI Models and BYOK

Q: What AI models are supported?

A: SpatialRead supports 300+ AI models including:

• OpenAI: GPT-5.1 Thinking, GPT-5.1 Instant, GPT-5 Pro, GPT-5, GPT-5 Mini, GPT-5 Nano
• Google: Gemini 3 Pro, Gemini 2.5 Pro, Gemini 2.5 Flash, Gemini 2.5 Flash Lite
• Perplexity: Sonar Pro, Sonar Reasoning Pro, Sonar Reasoning, Sonar
• Anthropic: Claude Opus 4.1, Claude 4.5
• OpenRouter: Access to 300+ models from various providers including Grok, Mistral, LLama, etc.
• Local models: Unlimited models via Ollama and LM Studio

Q: How does the Bring Your Own Keys (BYOK) model work?

A: Users provide their own API keys from AI providers (OpenAI, Anthropic, Google, Perplexity, etc.). AI queries go directly to the user's chosen provider—we never have access to user documents, API keys, or AI interactions. Users control their own costs and data.

Q: Are there any free AI options available?

A: Yes. Users can use Ollama to run local AI models completely free with no API costs. This provides unlimited, private AI usage without internet connectivity. LM Studio also offers free local model options.

Privacy and Data Collection

Q: Does SpatialRead track what students are reading?

A: No. We collect anonymous app performance metrics (like "PDF uploaded" or "AI response time") but never the content of PDFs, chat messages, annotations, or research topics. Your documents and notes stay entirely on your device. We track product health events, not research content.

Q: Where is analytics data stored?

A: Analytics data is processed by PostHog, which is SOC 2 Type II certified and GDPR/CCPA compliant. PostHog supports data residency controls and IP address anonymization. Data is used solely for product improvement and is never sold or shared with third parties.

Q: Can users opt out of analytics?

A: SpatialRead is a no-signup desktop application. Users can use the application completely offline with local models (Ollama) without any analytics collection. When using cloud AI models, analytics are limited to product health metrics and are anonymized.

Institutional Deployment

Q: Can SpatialRead be deployed institution-wide?

A: Yes. SpatialRead is a desktop application that can be distributed to students and researchers. Since it's a no-signup application with local data storage, there are no server requirements or institutional accounts needed. Each user downloads and uses the application independently. We offer a 20% student discount for individual purchases but if you're looking to deploy the paid version of the application institution-wide, please contact us at [email protected].

Q: Does SpatialRead comply with FERPA requirements?

A: SpatialRead is designed to be FERPA-friendly: it does not collect student data, personally identifiable information, or educational records. All documents and research work remain on the user's device. Anonymous usage analytics (if enabled) do not include student identifiers or educational content.

Q: Can SpatialRead be used for sensitive research projects?

A: Yes. SpatialRead's local-first architecture ensures that sensitive research documents never leave the user's device. When using local models (Ollama), all processing happens entirely offline with no data transmission. This makes it suitable for classified, proprietary, or highly sensitive research.

Q: What support is available for librarians and institutions?

A: We provide email support at [email protected] and typically respond within 48 hours. We're happy to provide additional information, materials for resource guides, or answer questions about institutional deployment, privacy, or technical requirements.

Use the following information when adding SpatialRead to your library's AI tools resource guide:

Basic Information

Tool Name: SpatialRead

Category: AI Research Tool / Research Assistant

Free/Paid: Free with local models via Ollama or paid for with Bring Your Own Keys (BYOK) model

Official Website: https://spatialread.com

Support Contact: [email protected]

Underlying Data/Model: User-provided API keys (OpenAI, Anthropic, Google, Perplexity, OpenRouter) or local models via Ollama and LM Studio

Short Description (50 words)

SpatialRead is a desktop application that helps students and researchers understand complex research papers through AI-powered explanations. Users upload PDFs, highlight confusing text, and receive contextual explanations. Features include spatial canvas organization, multi-model AI support, and complete privacy through local processing and BYOK architecture.

Detailed Description (150 words)

SpatialRead is a desktop application designed to accelerate research paper comprehension for students, researchers, and faculty. The tool allows users to upload research paper PDFs, highlight confusing text, and receive AI-powered explanations that are contextualized within the specific paper being read.

Key features include a spatial canvas for visual organization of ideas, support for 300+ AI models from major providers (OpenAI, Anthropic, Google, Perplexity, OpenRouter), and a highlight-to-explain workflow that makes dense academic content more accessible. The application uses a Bring Your Own Keys (BYOK) model, ensuring complete privacy—all documents and processing remain local on the user's device, with no data collection or cloud storage.

SpatialRead is particularly valuable for literature reviews, cross-disciplinary research, and situations where researchers need to quickly understand papers outside their primary expertise. Optional Ollama/LM Studio integration enables unlimited, free local AI usage without internet connectivity. Available for macOS and Windows.

We've prepared a complete package of materials to help you add SpatialRead to your resource guide:

Materials Package Includes

• Product Descriptions: Multiple length versions (10-word, 25-word, 50-word, 150-word)
• High-Resolution Screenshots: Application interface screenshots for your guide
• Logo Files: SpatialRead logo in multiple formats (SVG, PNG)
• Quick Reference Guide: One-page summary with feature list and technical specifications for easy reference

Download: SpatialRead Resources.zip

We're here to help you add SpatialRead to your library's resource guide. If you have questions about:

• Product features or capabilities
• Privacy and security architecture
• Technical requirements or compatibility
• Institutional licensing or bulk deployment
• Any other questions about SpatialRead

Please reach out to us at [email protected]. We typically respond within 48 hours and are happy to provide additional information or materials as needed.