[Feat] add PP Support for minimax-m2 series (#19577)

python/sglang/srt/models/minimax_m2.py

@@ -54,7 +54,7 @@ from sglang.srt.layers.moe.utils import get_moe_a2a_backend
 from sglang.srt.layers.quantization.base_config import QuantizationConfig
 from sglang.srt.layers.radix_attention import RadixAttention
 from sglang.srt.layers.rotary_embedding import get_rope
-from sglang.srt.layers.utils import PPMissingLayer
+from sglang.srt.layers.utils import PPMissingLayer, get_layer_id
 from sglang.srt.layers.vocab_parallel_embedding import (
     ParallelLMHead,
     VocabParallelEmbedding,
@@ -967,6 +967,7 @@ class MiniMaxM2ForCausalLM(nn.Module):
             self.lm_head = PPMissingLayer()
 
         self.logits_processor = LogitsProcessor(config)
+        self.pp_group = get_pp_group()
 
         # For EAGLE3
         self.capture_aux_hidden_states = False
@@ -999,17 +1000,26 @@ class MiniMaxM2ForCausalLM(nn.Module):
         positions: torch.Tensor,
         forward_batch: ForwardBatch,
         input_embeds: torch.Tensor = None,
+        pp_proxy_tensors: Optional[PPProxyTensors] = None,
     ) -> torch.Tensor:
-        # _print_tensor_info(input_ids, "input_ids")
+        hidden_states = self.model(
-        hidden_states = self.model(input_ids, positions, forward_batch, input_embeds)
+            input_ids,
+            positions,
+            forward_batch,
+            input_embeds,
+            pp_proxy_tensors=pp_proxy_tensors,
+        )
 
         aux_hidden_states = None
         if self.capture_aux_hidden_states:
             hidden_states, aux_hidden_states = hidden_states
 
+        if self.pp_group.is_last_rank:
             return self.logits_processor(
                 input_ids, hidden_states, self.lm_head, forward_batch, aux_hidden_states
             )
+        else:
+            return hidden_states
 
     def load_weights(self, weights: Iterable[Tuple[str, torch.Tensor]]):
         """Load model weights with proper mapping for MiniMax architecture."""
@@ -1038,6 +1048,17 @@ class MiniMaxM2ForCausalLM(nn.Module):
             if "rotary_emb.inv_freq" in name:
                 continue
 
+            layer_id = get_layer_id(name)
+            if (
+                layer_id is not None
+                and hasattr(self.model, "start_layer")
+                and (
+                    layer_id < self.model.start_layer
+                    or layer_id >= self.model.end_layer
+                )
+            ):
+                continue
+
             spec_layer = get_spec_layer_idx_from_weight_name(self.config, name)
             if spec_layer is not None:
                 continue  # skip spec decode layers for main model
@@ -1056,7 +1077,10 @@ class MiniMaxM2ForCausalLM(nn.Module):
                     continue
                 name = name.replace(weight_name, param_name)
                 # Skip loading extra bias for GPTQ models.
-                if name.endswith(".bias") and name not in params_dict:
+                if name not in params_dict:
+                    continue
+
+                if name.endswith(".bias"):
                     continue
 
                 param = params_dict[name]
@@ -1070,6 +1094,8 @@ class MiniMaxM2ForCausalLM(nn.Module):
                         continue
                     name = name.replace(weight_name, param_name)
 
+                    if name not in params_dict:
+                        continue
                     param = params_dict[name]
                     weight_loader = param.weight_loader
                     weight_loader(
@@ -1090,6 +1116,8 @@ class MiniMaxM2ForCausalLM(nn.Module):
                     if name is None:
                         continue
 
+                    if name not in params_dict:
+                        continue
                     param = params_dict[name]
                     weight_loader = getattr(
                         param, "weight_loader", default_weight_loader