Unriddle

Unriddle was an AI-powered research and document workspace designed to help students, researchers, academics, analysts, and professionals read, organize, understand, and write from complex information. The product has since been rebranded as Anara, with the company shifting its primary focus toward scientific research. The former Unriddle website now redirects to Anara, and existing Unriddle subscriptions were transferred to the Anara platform.

The platform is built around a source-grounded research workflow. Users can upload documents, organize them into folders, search across their research library, ask questions about individual files or groups of documents, and receive answers linked to the exact passages used. Anara currently says it can work across up to 10,000 files in a single conversation, allowing researchers to move from analyzing one document to synthesizing information across a much larger research collection.

One of the central capabilities is document understanding. The platform supports PDFs, Word documents, PowerPoint files, Excel spreadsheets, Markdown, plain text, EPUB, images, audio, video, webpages, and YouTube links. When a document is uploaded, Anara extracts metadata such as authors, DOI, and publication date and generates an AI summary in the background. This allows researchers to build a searchable library rather than repeatedly uploading the same files for individual questions.

Anara’s AI chat allows users to ask questions about a single paper, an entire folder, or their full research library. It can explain concepts, identify key findings, compare information between sources, and search through documents for specific evidence. The platform’s current system can also search the web and academic databases when those sources are enabled, giving researchers a way to combine their private document collection with external research.

A major differentiator is cited answers. Anara states that every answer points to the exact passage from the user’s sources, with clickable citations that let users inspect the underlying evidence. This is designed to make AI-assisted research more verifiable and to help users distinguish source-supported information from unsupported generated text.

The platform also contains a research library and organizational system. Users can create folders, add tags, annotate PDFs, take notes, save research material, and organize files into project-specific collections. This makes the product more than a PDF chatbot: it functions as a persistent research workspace in which documents and notes remain available for future questions.

Anara supports research writing directly within the same environment. Its note editor allows users to write while searching their library and inserting citations. The current product page says users can write with their library behind them and automatically match claims to relevant sources from their documents or the web. Documents can be exported to formats including PDF, Word, LaTeX, and HTML.

The platform also provides AI-assisted citation discovery. When researchers are writing notes or documents, Anara can identify relevant sources and suggest citations linked to the underlying research. Citation formats can be customized, with currently supported formats including numbered citations, APA 7th edition, MLA 9th edition, Harvard, Chicago, IEEE, CSE, ACS, Vancouver, ASA, AMA, and OSCOLA.

Another important feature is research-source discovery. Anara currently connects directly to academic databases such as PubMed, bioRxiv, medRxiv, ClinicalTrials.gov, Protein Data Bank, ChEMBL, and Open Targets. It can also search a database containing more than 240 million academic papers. This allows researchers to search for literature from within the same workspace where they store and analyze their own files.

Anara also supports Zotero and Mendeley. Researchers can import existing reference libraries into Anara, allowing their existing scholarly collections to become part of the AI research workspace. This is available across the current plans, while additional connectors are introduced at higher subscription levels.

The current platform has expanded its cloud and productivity connectors. Pro users can connect Google Drive, Notion, OneDrive, Dropbox, Slack, and Microsoft Teams, allowing the research workspace to work with information stored outside the Anara library. Enterprise plans add SharePoint live synchronization and broader connector access.

Anara also includes OCR and image understanding at higher subscription levels. Pro supports text extraction from images and scanned documents, making those materials searchable with AI. The Max plans extend this with semantic image recognition and broader research capabilities. These features are useful when important information is contained in scanned pages, figures, diagrams, or other non-standard document elements.

For complex research, the platform now offers Research and Deep Search agents. The Research agent is available across the product tiers, while Deep Search is an advanced Max feature designed for multi-step research tasks that explore the user’s library and the web in parallel. Max also includes Model Council, which allows a question to be posed to multiple AI models at once.

The service supports collaborative research through shared folders and team access. Plus allows up to five collaborators per folder, Pro increases this to 50, and Max provides unlimited collaborators. Enterprise adds administrative controls, certified security, SSO/SAML, and other organization-level features.

Security is a major part of the current product. Anara states that it is GDPR compliant, while its Enterprise environment adds SOC 2 Type II and ISO 27001 certification, an SSO/SAML option, custom user roles, user provisioning, and other enterprise controls. The company also states that Anara and its partner AI providers do not train models on customer data.

Anara is currently focused on scientific research, with the official website identifying academia, biopharma, clinical practice, engineering R&D, and policy as major workflow areas. The platform highlights applications such as thesis research, literature analysis, clinical research, regulatory work, engineering research, and evidence-backed policy work.

Language support is broad. The current Anara documentation states that its AI supports more than 90 languages, including French, Spanish, German, Hindi, and Chinese. The interface itself is currently available in English, while the AI can understand and respond in the larger multilingual set.

The original Unriddle product was launched by Naveed Janmohamed and began operating as a research assistant for academic and industry research. The company’s own 2025 announcement said it had been more than 18 months since Janmohamed wrote the first line of code, placing the project’s beginnings in 2023. Anara Labs, Inc. is now the company associated with the product.

The service currently uses a freemium subscription model. The Free plan costs $0 per seat and provides limited daily AI usage, uploads up to 20 MB or 120 pages, and Zotero and Mendeley connectors. Plus costs $10 per seat/month when billed yearly, Pro costs $20 per seat/month when billed yearly, and Max costs $100 per seat/month when billed yearly. Enterprise pricing is customized.

The pricing structure is now usage-based rather than simply counting document uploads. AI chat, text generation, extractions, and file imports contribute to a usage meter. Free usage refreshes daily, while paid plans refresh weekly. Paid users can purchase additional credits when they exceed their included usage.

Independent review data for Unriddle remains limited. G2 currently shows 0 reviews for Unriddle, so there is no reliable G2 star rating to report. Capterra likewise shows 0 user reviews for the former Unriddle listing, despite displaying historical pricing information. This means there is insufficient independent review data to establish a representative customer-satisfaction rating.

Unriddle, now Anara, is best suited to users who need a research-focused workspace that combines document understanding, source-grounded AI, literature discovery, citation support, note-taking, and collaboration. Its current direction is substantially more specialized toward scientific research than the earlier Unriddle product, making the rebrand an important part of understanding what the service offers today.

Pricing

• Free: $0/month per seat when billed yearly, with limited AI usage, uploads up to 20 MB or 120 pages, and Zotero/Mendeley connectors.
• Plus: $10/month per seat when billed yearly.
• Pro: $20/month per seat when billed yearly, including broader AI-model access, OCR, and cloud/productivity connectors.
• Max: $100/month per seat when billed yearly, including Deep Search and Model Council.
• Enterprise: Custom pricing.

Review

Unriddle currently has 0 G2 reviews and 0 Capterra user reviews, so there is not enough independent review data to provide a meaningful star rating. The current product has instead evolved into Anara, whose public positioning centers on scientific research, source-grounded answers, document analysis, and research collaboration.