Mistral Large 4 searches the web through SearchBase with one tool entry, "engine": "searchbase". Mistral Large 4 has no search of its own, so this is how it reads the web.
SearchBasewithinPaste this into your coding agent (Claude Code, Cursor, Codex, and more) and it will set up and build with Opper for you.
curl https://api.opper.ai/v3/compat/chat/completions \-H "Authorization: Bearer $OPPER_API_KEY" \-H "Content-Type: application/json" \-d '{"model": "mistral:eu/mistral-large-4","messages": [{"role": "user", "content": "What changed in the news today?"}],"tools": [{"type": "opper:web_search", "engine": "searchbase"}]}'
Mistral Large 4 runs SearchBase on the routes below.
Mistral Large 4, Mistral AI's flagship open-weight model: a 1T-parameter mixture of experts (49B active) with native image input, tool use, optional reasoning and long context.
Step-by-step setup for SearchBase on each of these models through Opper.
Opper's default search engine for models without their own. Search processing stays in the EU.
"engine": "jina"
Exa's search engine built for AI, on any tool-calling model through one tool entry. US-based.
"engine": "exa"
Linkup's search for AI apps, running in the US, EU, Canada and APAC, on any tool-calling model.
"engine": "linkup"