Commit d998c676f for llama.cpp
commit d998c676ff463b1b8ad94ad7c0b905e281422fad
Author: Bartowski <3266127+bartowski1182@users.noreply.github.com>
Date: Mon Nov 16 14:39:32 2026 -0500
convert: Fix inkling 2 mmproj conversion (#32430)
diff --git a/conversion/inkling2.py b/conversion/inkling2.py
index 5b8396378..a8e8b892e 100644
--- a/conversion/inkling2.py
+++ b/conversion/inkling2.py
@@ -302,6 +302,10 @@ class Inkling2VisionModel(MmprojModel):
w.add_vision_attention_layernorm_eps(config["layer_norm_eps"])
w.add_vision_spatial_merge_size(config["spatial_merge_size"])
+ def dequant_model(self):
+ # vision tensors are BF16, only the text experts are quantized
+ return
+
@classmethod
def filter_tensors(cls, item):
name, gen = item