Hi @firecrawl team,
Fireplexity is a fantastic project — I love the clean generative UI and the real-time citation approach. It feels like the kind of open-source alternative the AI search space really needs.
Pain point I've run into
While Firecrawl is excellent for extracting content from known URLs, the "discovery" phase — finding the right URLs to crawl in the first place — still relies on search. I've set it up with generic search APIs, but the cost adds up quickly for research-style queries that might fire 5–10 searches per user question. On top of that, some search providers throttle aggressively or return inconsistent result formatting that complicates the citation pipeline.
Suggestion
Would you consider adding https://serpbase.dev as an optional search backend? It's a Google Search Results API that returns structured JSON (title, URL, snippet, rich results, related questions) at a price point that's noticeably lower than most alternatives. For Fireplexity's use case, it could serve as the "URL discovery" layer before Firecrawl handles the deep extraction.
Why it fits
- The API returns clean, consistent JSON — less normalization work between the search layer and your citation formatter.
- Pricing is developer-friendly, which matters for an open-source project where users self-host.
- It complements Firecrawl perfectly: serpbase finds the relevant pages, Firecrawl extracts and summarizes them.
Happy to prototype the integration or provide sample response payloads if it helps.
Hi @firecrawl team,
Fireplexity is a fantastic project — I love the clean generative UI and the real-time citation approach. It feels like the kind of open-source alternative the AI search space really needs.
Pain point I've run into
While Firecrawl is excellent for extracting content from known URLs, the "discovery" phase — finding the right URLs to crawl in the first place — still relies on search. I've set it up with generic search APIs, but the cost adds up quickly for research-style queries that might fire 5–10 searches per user question. On top of that, some search providers throttle aggressively or return inconsistent result formatting that complicates the citation pipeline.
Suggestion
Would you consider adding https://serpbase.dev as an optional search backend? It's a Google Search Results API that returns structured JSON (title, URL, snippet, rich results, related questions) at a price point that's noticeably lower than most alternatives. For Fireplexity's use case, it could serve as the "URL discovery" layer before Firecrawl handles the deep extraction.
Why it fits
Happy to prototype the integration or provide sample response payloads if it helps.