fix: gemma 3 not use softcap (sgl-project#5622)

Author: zhyncs

python/sglang/srt/configs/model_config.py

@@ -78,6 +78,11 @@ def __init__(
                 logger.info(
                     "Multimodal is disabled for Llama4. To enable it, set --enable-llama4-multimodal."
                 )
+            elif self.hf_config.architectures[0] == "Gemma3ForConditionalGeneration":
+                enable_multimodal = False
+                logger.info(
+                    "Multimodal is disabled for Gemma3. To enable it, set --enable-gemma3-multimodal."
+                )
             else:
                 enable_multimodal = True
 

python/sglang/srt/models/gemma3_causal.py

@@ -189,7 +189,7 @@ def __init__(
             self.scaling,
             num_kv_heads=self.num_kv_heads,
             layer_id=layer_id,
-            logit_cap=getattr(self.config, "attn_logit_softcapping", None),
+            logit_cap=0.0,
             # Module must also define `get_attention_sliding_window_size` to correctly initialize
             # attention backend in `ForwardBatch`.
             sliding_window_size=self.sliding_window,

python/sglang/srt/server_args.py

@@ -154,6 +154,7 @@ class ServerArgs:
     disable_outlines_disk_cache: bool = False
     disable_custom_all_reduce: bool = False
     enable_llama4_multimodal: Optional[bool] = None
+    enable_gemma3_multimodal: Optional[bool] = None
     disable_overlap_schedule: bool = False
     enable_mixed_chunk: bool = False
     enable_dp_attention: bool = False
@@ -285,7 +286,9 @@ def __post_init__(self):
         if self.grammar_backend is None:
             self.grammar_backend = "xgrammar"
 
-        self.enable_multimodal: Optional[bool] = self.enable_llama4_multimodal
+        self.enable_multimodal: Optional[bool] = (
+            self.enable_llama4_multimodal or self.enable_gemma3_multimodal
+        )
 
         # Data parallelism attention
         if self.enable_dp_attention:
@@ -984,6 +987,12 @@ def add_cli_args(parser: argparse.ArgumentParser):
             action="store_true",
             help="Enable the multimodal functionality for Llama-4.",
         )
+        parser.add_argument(
+            "--enable-gemma3-multimodal",
+            default=ServerArgs.enable_gemma3_multimodal,
+            action="store_true",
+            help="Enable the multimodal functionality for Gemma-3.",
+        )
         parser.add_argument(
             "--disable-overlap-schedule",
             action="store_true",

python/sglang/srt/utils.py

@@ -1971,6 +1971,7 @@ def is_fa3_default_architecture(hf_config):
         "LlamaForCausalLM",
         "MistralForCausalLM",
         "Gemma2ForCausalLM",
+        "Gemma3ForConditionalGeneration",
     }
     return architectures[0] in default_archs