Markdown rendering
Every assistant turn is rendered as GitHub-flavoured Markdown as the response streams. You see formatted output the moment the tokens arrive, not after the model finishes.
What works out of the box
- Headings, lists, blockquotes — standard CommonMark.
- Tables — GitHub-style pipe tables with alignment.
- Task lists —
- [ ]and- [x]render as checkboxes. - Footnotes — inline reference markers with linked-back definitions.
- Math — LaTeX-style inline
$\dots$and display$$\dots$$expressions. - Fenced code — language-aware syntax highlighting, plus a toolbar for copy, export and open in canvas.
Copying and selecting
Right-click any assistant message for a menu with Copy raw (the original markdown source) and Copy as markdown (the rendered text serialized back to markdown). You can also highlight any range of text and a small Copy selection pill will appear at the cursor.
Long user messages
User bubbles that paste in dozens of lines automatically clamp to the first ten lines with a Show more toggle, so the assistant’s reply does not get pushed off-screen by your own input.
Regenerate the last reply
The last assistant message in a chat carries a Regenerate action. Triggering it drops the existing reply, re-sends the preceding user turn, and streams a fresh answer in place — useful when the model wanders off, hallucinates, or you just want a second take without retyping the prompt.
Only the most recent assistant message can be regenerated. Earlier turns stay intact so the rest of the conversation history is preserved.
Thinking traces
For reasoning-capable models, the chain-of-thought stream is rendered into a separate, collapsible block above the answer. You can fold it away when you only care about the final answer and pop it back open to audit how the model got there.