Berges WebSearch
Search and fetch when the answer needs primary sources. Pick WebSearch from the sidebar; the model decides per question whether to look, and every search shows on the reply.
WebSearch is the pipeline for questions that need fresh sources. Pick it from the sidebar and the deep-thinking layer gains two tools: web_search for finding candidate sources and fetch_url for reading them in full. The model decides when to use them.
Selecting it does not mean browsing on every turn: the model searches only when a question calls for current information or a primary source, and every search and fetch shows as a badge on the reply. Switch to Reasoning, Brainstorm or Empathy when you want a chat that never leaves the model.
What changes when you pick it
Two tools
web_search returns title + URL + snippet for a query. fetch_url returns the main text of a specific page. The model typically searches first, then fetches the best result.
Iterative
A single message can include multiple searches and fetches. The model loops through tool calls until it has what it needs, then writes the final answer with the URLs it used.
Visible activity
Every search and fetch shows up as a badge in your message bubble ("Searched: X", "Fetched Y"). You can tell at a glance whether the answer came from the web or just the model.
Runs on every turn
While WebSearch is selected the tools are available on every reply, and the model decides whether a given question needs them. Pick Reasoning, Brainstorm or Empathy when you want chats fully offline.
When WebSearch is the right pick
- → Current events, recent news, or anything that depends on this week's state of the world
- → Named entities the model might not know (new products, niche companies, recent papers)
- → Pricing, specifications, or release dates that change over time
- → Anything you'd want to cite a source for
When to skip it
- × Casual chat or brainstorming. The WebSearch pipeline adds latency and discloses your queries to a third-party search provider.
- × Sensitive topics you don't want a search provider to see. The Brainstorm or Reasoning pipelines stay fully offline.
- × Real-time data (live scores, stock ticks). Search results are minutes-to-hours stale; we're not a market data feed.
Things people ask about WebSearch
Who sees my search queries?
When WebSearch is active, your search queries are sent to a third-party search provider (currently Brave Search). They process the query to return results and do not retain it for training. See the Privacy Policy for details.
Will WebSearch make my chats slower?
Yes, when it actually uses the tools. Each search and fetch adds a round trip. A typical WebSearch-assisted answer takes a few extra seconds compared to a non-WebSearch answer.
Can the model fetch any URL?
Almost. We block private, loopback, and link-local IP ranges at the network level so the model can't poke at internal services. Public URLs work; size and time limits apply.
Do you store the search results?
No. Results are fed to the model for the turn and discarded. Your message and the final assistant reply are kept only if you save the conversation (encrypted at rest, same as everything else).