diff --git a/python/sglang/srt/models/commandr.py b/python/sglang/srt/models/commandr.py
index 151087732..6d2e6d2bb 100644
--- a/python/sglang/srt/models/commandr.py
+++ b/python/sglang/srt/models/commandr.py
@@ -61,7 +61,10 @@ from sglang.srt.layers.radix_attention import RadixAttention
 from sglang.srt.layers.rotary_embedding import get_rope
 from sglang.srt.layers.vocab_parallel_embedding import VocabParallelEmbedding
 from sglang.srt.model_executor.forward_batch_info import ForwardBatch
-from sglang.srt.model_loader.weight_utils import default_weight_loader
+from sglang.srt.model_loader.weight_utils import (
+    default_weight_loader,
+    maybe_remap_kv_scale_name,
+)
 from sglang.srt.utils import get_compiler_backend, set_weight_attrs
 
 
@@ -372,6 +375,11 @@ class CohereForCausalLM(nn.Module):
                 # Skip loading extra bias for GPTQ models.
                 if name.endswith(".bias") and name not in params_dict:
                     continue
+                # Remapping the name of FP8 kv-scale.
+                name = maybe_remap_kv_scale_name(name, params_dict)
+                if name is None:
+                    continue
+
                 param = params_dict[name]
                 weight_loader = getattr(param, "weight_loader", default_weight_loader)
                 weight_loader(param, loaded_weight)
diff --git a/python/sglang/srt/models/dbrx.py b/python/sglang/srt/models/dbrx.py
index cedc96392..92fc67939 100644
--- a/python/sglang/srt/models/dbrx.py
+++ b/python/sglang/srt/models/dbrx.py
@@ -42,7 +42,10 @@ from sglang.srt.layers.vocab_parallel_embedding import (
     VocabParallelEmbedding,
 )
 from sglang.srt.model_executor.forward_batch_info import ForwardBatch
-from sglang.srt.model_loader.weight_utils import default_weight_loader
+from sglang.srt.model_loader.weight_utils import (
+    default_weight_loader,
+    maybe_remap_kv_scale_name,
+)
 from sglang.srt.utils import set_weight_attrs
 
 
@@ -411,6 +414,11 @@ class DbrxForCausalLM(nn.Module):
                 weight_loader(param, loaded_weight, weight_name)
                 break
             else:
+                # Remapping the name of FP8 kv-scale.
+                name = maybe_remap_kv_scale_name(name, params_dict)
+                if name is None:
+                    continue
+
                 param = params_dict[name]
                 weight_loader = getattr(param, "weight_loader", default_weight_loader)
                 weight_loader(param, loaded_weight)