Pick Perplexica if
people who will run Docker and an LLM (local or BYOK) to keep search+answer off a hosted Pro plan.
Read the Perplexica pagePerplexica vs Perplexity
Our take
Pick Perplexica if people who will run Docker and an LLM (local or BYOK) to keep search+answer off a hosted Pro plan. Pick Perplexity if researchers, analysts, and anyone who wants a sourced first pass before opening ten tabs. Skip Perplexica if someone who wants a signed-out hosted composer with a published $17 Pro card (Perplexity) and will not operate search + model. Skip Perplexity if teams that cannot send
Perplexica
Self-host AI search (now Vane). MIT
Best for
Perplexica Perplexity-shaped search on a box you run
Perplexity Research with sources you can click
Free plan
Perplexica $0 · mIT. You pay the LLM (local GPU or an API key) and the search host
Perplexity $0 · standard Search on the public composer
Lowest paid
Perplexica Not published by the vendor
Perplexity Pro $17/mo yearly
Next step up
Perplexica
Perplexity Max from $167/mo yearly
Coding agent
Perplexica
Perplexity API platform
Runs where
Perplexica Local or hosted
Perplexity Hosted
Watch out
Perplexica You operate two systems
Perplexity Citations are not a guarantee
Last checked
Perplexica 2026-10-20 · github.com/ItzCrazyKns/Vane
Perplexity 2026-10-19 · perplexity.ai/pro
Which one, then?
Pick Perplexity if
researchers, analysts, and anyone who wants a sourced first pass before opening ten tabs.
Read the Perplexity page