AI Potluck
Contribute
Back to Gap Map Product / UX / UI & API

Perplexity

Perplexity AI
closed / Overall score: 4.7(leading)

Search-grounded consumer chat product that answers questions with cited sources. It routes over models from several providers including OpenAI, Anthropic and Meta, and adds guided deep research, a voice mode, and an assistant that drafts messages and makes bookings. It ships as a mobile app that syncs across devices, distinct from the Perplexity API.

perplexity.ai serves a client-rendered shell, so the consumer surface is described from Perplexity's own App Store listing rather than its website.

Openness

1 high confidence
1.0
license
proprietary(SaaS)
source
closed
self-host
none
models
multi-provider(closed orchestration)

Perplexity is a proprietary SaaS product: its own developer documentation sells only hosted API access to models it runs on its own infrastructure, with no repository, no self-hostable build and no license grant anywhere. That documentation describes the API rather than the consumer app directly, but nothing published for either surface offers source or self-hosting.

  • https://docs.perplexity.ai/ recorded 2026-09-29

    Perplexity's own API docs, which offer only a hosted service: "Unified access to open-weight models hosted by Perplexity through one OpenAI-compatible endpoint" and "access third-party models from OpenAI, Anthropic, Google, and xAI with web search tools and presets". Searched the body for self-host, on-prem, open source and license: none.

  • https://docs.perplexity.ai/docs/resources/faq.md recorded 2026-09-29

    Perplexity's own FAQ on where the models run: "Our compute is hosted via Amazon Web Services in North America." The models run on Perplexity's infrastructure and are reached over its API.

  • https://www.perplexity.ai/ recorded 2026-06-04

    consumer AI answer/search chat product, proprietary SaaS (no source/self-host)

Adoption

5 medium confidence
5.0

Perplexity does not publish its own usage figures on any reachable page, so the reported reach comes from a Business of Apps compilation covering its website and app rather than from first-party numbers.

Capability

4 medium confidence
4.0

A differentiated and capable consumer chat interface - citation-grounded answers and model choice - with high feature breadth, though it does not define the category frontier on its own. Multimodal input and Spaces or collections could not be confirmed on any reachable page, so neither is claimed here; the remaining surface still supports a 4 on the same reasoning.

  • https://docs.perplexity.ai/ recorded 2026-09-29

    Model choice and grounding from the vendor's own docs: "web-grounded answers with built-in citations in one call" and "access third-party models from OpenAI, Anthropic, Google, and xAI with web search tools and presets".

  • https://docs.perplexity.ai/docs/agent-api/migrate-from-sonar/overview.md recorded 2026-09-29

    Migration guide: "Full flexibility to select across the OpenAI, Anthropic, Google, xAI, Z.AI, Moonshot AI, and NVIDIA model families, in addition to Perplexity's Sonar model." Also states "Sonar Chat Completions support ended on September 27, 2026." and lists web search, URL fetching, sandboxes and MCP as built-in tools.

  • https://docs.perplexity.ai/docs/agent-api/presets.md recorded 2026-09-29

    Presets page: "Presets are pre-configured setups optimized for specific use cases." and "Each preset bundles a model, search config, reasoning steps, system prompt, and available tools."

  • https://en.wikipedia.org/wiki/Perplexity_AI recorded 2026-09-29

    Paid tier depth: Internal Knowledge Search "enables Pro and Enterprise Pro users to browse web content and their internal documents simultaneously" and lets users upload and search files. Also lists the selectable frontier models in Pro.

  • https://www.perplexity.ai/ recorded 2026-06-04

    multi-model selection, web-grounded cited answers, file/multimodal features

Verified 2026-09-29