[model] support MiniCPM-V-4.6 (#10472)

src/llamafactory/data/collator.py

@@ -17,6 +17,7 @@
 
 import copy
 import inspect
+import os
 from dataclasses import dataclass
 from typing import TYPE_CHECKING, Any, Literal, Optional
 
@@ -474,6 +475,13 @@ class MultiModalDataCollatorForSeq2Seq(DataCollatorForSeq2Seq):
             features["position_ids"] = torch.arange(seq_length).long().repeat(bsz, 1)
             return {"data": features, "input_ids": features["input_ids"], "labels": features["labels"]}
 
+        if (
+            self.model is not None
+            and getattr(self.model.config, "model_type", None) == "minicpmv4_6"
+            and "target_sizes" in features
+        ):  # for minicpmv4_6 with new transformers (NaViT API, no image_bound)
+            features["downsample_mode"] = os.getenv("DOWNSAMPLE_MODE", "16x")
+
         return features