As of June 2026 — menus move. Help: Claude support · Research.
Open claude.ai and sign in.
1 — Find the model picker
- Start or open a chat.
- Find the model name at the top (e.g. Sonnet, Opus, Haiku — labels vary by plan).
- Note what's selectable on your account.
2 — Find Research or web search
Pick one path:
| Path | Steps |
|---|---|
| Research | Left sidebar or + new → Research (multi-step reports) |
| Web search | In chat composer → + → Web search (or enable search in settings) |
Open the screen. You don't need to finish a full research job — find the entry point.
3 — Find Artifacts or Projects
| Feature | Where to look |
|---|---|
| Artifacts | Send a prompt that asks for a doc/table — side panel may open; or mentioned in help |
| Projects | Left sidebar → Projects |
Skim one — know it exists for later lessons.
4 — Confirm: no image gen in main chat
Look for Create image or DALL·E in the main Claude chat UI.
Expected: you won't find native image generation here (that's the ChatGPT/Gemini contrast from Lesson 7). If you do find it, update your cheat sheet — products change.
You should see: model picker, Research or web search entry, Projects or Artifacts awareness, and a note on images: yes/no in this app.
Claude is where I go the moment ChatGPT starts sounding like a motivational poster. But I still reach for ChatGPT when I want a quick image mock-up — different tools for different jobs, not "which AI is cleverer."
Continue — Gemini.