Answers from your own data.
Upload PDFs, Word docs, spreadsheets, and text. Ask questions. Get answers grounded in your sources with inline citations.
“Per the Q3 OKR review (page 12), the marketing team committed to 200 net-new SQLs by end of quarter.” — Wirebase, citing your own document.
Drop in your documents.
Upload PDFs, Word documents, spreadsheets, Markdown, and plain text — Wirebase chunks and embeds them automatically using your chosen embedding provider. Knowledge bases are scoped per organization so each team's data stays separate, and when a document changes you can re-index it on demand.
- PDF, Word, Excel, Markdown, plain text
- Automatic chunking and embedding
- Per-organization scoping
- Re-index any document on demand
Your data, your database.
Vector search runs inside your existing PostgreSQL via pgvector — there's no third-party vector database to provision, pay for, or worry about. Bring your own embedding provider (OpenAI, Google, Cohere, or a local Ollama model), and data isolation is enforced at the row level so one organization can never retrieve another's documents. Self-host it and your documents never leave your infrastructure.
- pgvector inside your existing PostgreSQL
- Embedding provider of your choice: OpenAI, Google, Cohere, or local Ollama
- Hybrid retrieval: vector similarity plus full-text search
- Per-org isolation enforced at the row level
Trust, but verify.
Every answer includes inline citation markers that point back to the exact chunks the model used, and a Sources list under the reply spells out each one — filename, similarity score, and a snippet of the retrieved text. Click a marker to highlight its entry in that list. When the model's answer doesn't match the source, you'll know immediately — citations make hallucinations visible.
- Inline citation markers in every answer
- Click a marker to highlight that source in the Sources list
- Filename, similarity score, and text snippet for each source
- Hybrid search: semantic similarity plus keyword match
Retrieval that holds up.
Document types
PDF to spreadsheet, all chunked consistently.
Hybrid retrieval
pgvector similarity plus full-text search.
Org isolation
Knowledge never crosses organization boundaries.
Re-index on demand
Reprocess any document after it changes.
Cited answers
Inline markers tied to source chunks.
Embedding choice
Bring your own embedding provider.