SearXNG hosting
A private metasearch engine — and the search backend your agents need.
SearXNG hosting at AIHostBox means a dedicated, single-tenant SearXNG instance on our infrastructure: we install it, pin it to a reviewed release, put it behind TLS on your domain, back it up nightly and keep it patched. Plans start at $5.99 per month and the admin account, the data and any API keys are yours.
SearXNG queries dozens of search engines and returns aggregated results without profiling you, storing your queries or serving ads. As a personal search engine that is reason enough; as infrastructure it is more interesting still, because it gives your AI stack a search API that costs a flat monthly fee instead of per query.
Open WebUI, Perplexica, Dify and most agent frameworks can use a SearXNG instance for web search. Running your own means those searches leave from your instance rather than from a commercial API tied to your account.
SearXNG hosting plans
Every plan is a dedicated single-tenant instance with NVMe storage, nightly off-box backups, TLS and managed patching. Quarterly, semi-annual and annual terms are discounted at checkout. Model tokens are not included — you bring your own key.
How we run SearXNG
The official SearXNG container serving on port 8080 behind TLS, with Valkey alongside for the bot-detection limiter, configuration in /etc/searxng/settings.yml and a persistent cache volume.
What we configure for you
- SEARXNG_SECRET set to a random value
- SEARXNG_BASE_URL matching your hostname
- json added to search.formats for agent and chat front-end access
- The limiter enabled, with SEARXNG_VALKEY_URL pointing at the Valkey container
- Default engines, safe_search level and autocomplete backend chosen with you
Which plan fits
Starter (2 GB, 1 vCPU) is plenty for one person plus a chat front-end or agent issuing searches. Plus (4 GB, 2 vCPU) suits a team or a busy agent pipeline that fans many queries out in parallel. Pro (8 GB, 3 vCPU) is rarely needed: the limit you meet first is upstream engines throttling one IP, not memory.
The honest limits
Result quality depends on engines you do not control, and a single IP sending many automated queries will see engines suspended or challenged at times, so SearXNG is no substitute for a paid search API with guaranteed volume.
What people run SearXNG for
Search backend for AI
Give agents and chat UIs web access without per-query pricing.
Private search engine
Aggregate results without a profile following you.
Research automation
Programmatic queries through the JSON API.
Licence and version policy
| Item | Position |
|---|---|
| Upstream licence | AGPL-3.0 |
| Sold as managed hosting | Yes |
| Tenancy | Single tenant — one customer per instance |
| Patching | Critical advisories within 24 hours, high severity within 72 |
Our full position on licences, including the applications we decline to sell, is on the licensing page.
Works well with SearXNG
Open WebUI
A ChatGPT-style workspace for your team, pointed at any model you like.
from $5.99/moDetails →
Perplexica
An open answer engine — searches the web, then explains with citations.
from $11.99/moDetails →
OpenClaw
Your always-on personal AI assistant, running 24/7 on a server you control.
from $9.99/moDetails →
Dify
A visual studio for building LLM apps, agents and RAG pipelines.
from $18.99/moDetails →More Search & web agents we host
SearXNG hosting: your questions
Is the JSON API enabled for AI tools?
Will search engines block my instance?
How is the instance protected from bots and abuse?
Which search engines can I enable?
What does the AGPL-3.0 licence mean for me?
Run SearXNG without running a server
A dedicated SearXNG instance, deployed and maintained by us. Month to month, cancel whenever.