Put two characters in one image

tendre.AI can place two trained characters in the same scene while keeping both faces recognizable. This needs two separate LoRA, one per character, loaded into the same generation.

Train both characters first

Each character needs its own LoRA, trained the normal way from its own photo dataset. Do not try to mix two people into one LoRA, they need to be trained and stored separately for a duo to work.

Load both LoRA together

In the generation panel, add both LoRA to the same render. tendre.AI blends the two identities so each one applies to a distinct figure in the composition instead of merging into one face.

Describe who is where

Be explicit in the prompt about position and action for each character (left, right, facing each other, sitting together). The clearer the description, the more reliably each LoRA lands on the intended figure.

Balance the weights

If one character comes through stronger than the other, lower its LoRA weight slightly and raise the other's. Small adjustments, around 10 to 20 percent, are usually enough to even out a duo shot.

When it struggles

Very close framing or overlapping poses are harder to keep separate; a little distance between the two figures in the prompt makes the composition easier to render correctly. For the single-character basics, see Character consistency and Train a LoRA.