From 7e1370694f18f03cd6e6b84330c9531c6295ff38 Mon Sep 17 00:00:00 2001
From: Xuan Son Nguyen <son@huggingface.co>
Date: Fri, 7 Aug 2026 15:38:49 +0200
Subject: [PATCH] nits

---
 tools/mtmd/clip.cpp       | 2 +-
 tools/mtmd/mtmd-image.cpp | 6 +++---
 2 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/tools/mtmd/clip.cpp b/tools/mtmd/clip.cpp
index 51859cd8c6..17e60b6ba2 100644
--- a/tools/mtmd/clip.cpp
+++ b/tools/mtmd/clip.cpp
@@ -1596,7 +1596,7 @@ struct clip_model_loader {
                         if (hparams.image_longest_edge == 0) {
                             hparams.image_longest_edge = 3024;
                         }
-                        hparams.set_limit_image_tokens();
+                        // note: the step3vl preprocessor slices based on a fixed window grid, so it does not support custom min/max image tokens
                         hparams.warmup_image_size = hparams.image_size;
                     } break;
                 case PROJECTOR_TYPE_YOUTUVL:
diff --git a/tools/mtmd/mtmd-image.cpp b/tools/mtmd/mtmd-image.cpp
index 6f2a3784cd..073d83d453 100644
--- a/tools/mtmd/mtmd-image.cpp
+++ b/tools/mtmd/mtmd-image.cpp
@@ -141,9 +141,9 @@ struct img_tool {
 
     struct calc_size_opt {
         int align_size = 1;
-        int min_pixels = 0;
-        int max_pixels = 0;
-        // if longest_edge > 0, both longest_edge and min/max_pixels will be respected
+        int min_pixels = 0;   // 0 = disabled
+        int max_pixels = 0;   // 0 = disabled
+        // applied before min/max_pixels, so min_pixels can push an edge back above longest_edge
         int longest_edge = 0; // 0 = disabled
     };