What Claude does with a Markdown file, and where an image fits
Claude takes Markdown through a message attachment, or the knowledge base of a Project, and that is the door this page is about. The question worth asking is what an image looks like by the time it arrives, because Claude never sees the image itself. It sees whatever the conversion handed over, and nothing else.
Claude reads a fenced code block as something not to paraphrase and a heading as a citable section. What comes out of an image here is the text of the image as Markdown, with a table rebuilt when the image held one, so the structure Claude is looking for is actually present instead of merely implied. That is the entire trick, and there is nothing clever underneath it.
What survives from an image into Claude
The conversion drops the background, the stamps, the shadows and the skew of the scan before Claude ever sees them, because none of it carries meaning into Claude and all of it costs something. What it keeps is the part you would otherwise have retyped by hand.
- The text comes out in reading order rather than in scan order, and Claude receives it as structure rather than as something to infer.
- A visible table is rebuilt as a Markdown table instead of collapsing into a line of words, which is the part Claude would otherwise have to reconstruct on its own.
- References and figures are read character by character, which is the part people retype by hand, so nothing downstream of Claude has to guess at it.
What Claude never gets back from an image
Every conversion costs something, and a page listing only the gains is a sales page. Here is what an image gives up on the way to Claude, stated plainly enough that you can decide against it.
- Layout and colour are gone, and a colour that meant something meant it only to a human, and Claude will not get it back from anywhere.
- Anything too small or too blurred to read is not returned, and not flagged either, so nothing you build in Claude should depend on it.
- Certainty is gone: a model is reading, and a model can be wrong while sounding sure, which is a real cost inside Claude and not a rounding error.
The Image to Markdown step behind this Claude page was measured
Gemini 2.5 Flash Lite does the conversion, and it was proven end to end on 30 August 2026: a purchase order image built for the occasion, carrying a unique reference, came back with the reference, the quantity and the unit price all read correctly. That is one run on one day and not an average, which is exactly why the date travels with the claim instead of being left out of it. The same output is what Claude would have received.
This capability returned nothing at all until 30 August 2026, and two probes failed that same day before the quota was restored. Claude has no way of telling you any of that from the inside, so it gets said here instead, on the page that sent you there.
Once the image is Markdown, what to do inside Claude
With the Markdown in hand, load the Markdown into a Project once, so every later conversation starts from it instead of re-uploading. That is a Claude habit rather than a conversion setting, and it is where most of the value of converting an image in the first place actually lands.
The constraint worth knowing before you start: a window of 200k tokens is large enough to hide the exact moment a document stopped fitting inside it. It bites the same way whether the Markdown came out of an image or out of something else entirely, but it bites hardest on long documents, which is what an image usually is.
What we will not pretend about Image in Claude
- Claude will not re-read an attachment you replaced unless you say so, and converting an image first does not change that in the slightest.
- long context rewards structure and punishes filler, so extracted layout costs accuracy twice over, which is a Claude behaviour rather than anything the image conversion did.
- The conversion costs half a credit and the price is shown before the image is sent, whether the Markdown ends up in Claude or nowhere at all.