[feat] Support EAGLE3 for Qwen2 (#9216)

python/sglang/srt/models/qwen2.py

@@ -16,7 +16,7 @@
 # Modify details for the adaptation of Qwen2 model.
 """Inference-only Qwen2 model compatible with HuggingFace weights."""
 import logging
-from typing import Any, Dict, Iterable, Optional, Tuple, Union
+from typing import Any, Dict, Iterable, List, Optional, Tuple, Union
 
 import torch
 from torch import nn
@@ -431,7 +431,6 @@ class Qwen2ForCausalLM(nn.Module):
                     quant_config=quant_config,
                     prefix=add_prefix("lm_head", prefix),
                 )
-
         else:
             # ranks other than the last rank will have a placeholder layer
             self.lm_head = PPMissingLayer()
@@ -452,6 +451,8 @@ class Qwen2ForCausalLM(nn.Module):
 
         self.logits_processor = LogitsProcessor(config)
         self.pooler = Pooler(pooling_type=PoolingType.LAST, normalize=True)
+        # For EAGLE3 support
+        self.capture_aux_hidden_states = False
 
     def get_input_embedding(self, input_ids: torch.Tensor) -> torch.Tensor:
         return self.model.get_input_embedding(input_ids)
@@ -476,11 +477,18 @@ class Qwen2ForCausalLM(nn.Module):
             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:
             if not get_embedding:
                 return self.logits_processor(
-                    input_ids, hidden_states, self.lm_head, forward_batch
+                    input_ids,
+                    hidden_states,
+                    self.lm_head,
+                    forward_batch,
+                    aux_hidden_states,
                 )
             else:
                 return self.pooler(hidden_states, forward_batch)
@@ -619,5 +627,20 @@ class Qwen2ForCausalLM(nn.Module):
     def load_kv_cache_scales(self, quantization_param_path: str) -> None:
         self.model.load_kv_cache_scales(quantization_param_path)
 
+    def set_eagle3_layers_to_capture(self, layer_ids: Optional[List[int]] = None):
+        if not self.pp_group.is_last_rank:
+            return
+
+        self.capture_aux_hidden_states = True
+        if layer_ids is None:
+            num_layers = self.config.num_hidden_layers
+            self.model.layers_to_capture = [
+                2,
+                num_layers // 2,
+                num_layers - 3,
+            ]  # Specific layers for EAGLE3 support
+        else:
+            self.model.layers_to_capture = [val + 1 for val in layer_ids]
+
 
 EntryClass = Qwen2ForCausalLM

python/sglang/srt/models/qwen2_moe.py

@@ -17,7 +17,7 @@
 """Inference-only Qwen2MoE model compatible with HuggingFace weights."""
 
 import logging
-from typing import Any, Dict, Iterable, Optional, Tuple, Union
+from typing import Any, Dict, Iterable, List, Optional, Tuple, Union
 
 import torch
 import torch.nn.functional as F
@@ -536,6 +536,8 @@ class Qwen2MoeForCausalLM(nn.Module):
             use_attn_tp_group=global_server_args_dict["enable_dp_lm_head"],
         )
         self.logits_processor = LogitsProcessor(config)
+        # For EAGLE3 support
+        self.capture_aux_hidden_states = False
 
     @torch.no_grad()
     def forward(
@@ -553,9 +555,12 @@ class Qwen2MoeForCausalLM(nn.Module):
             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
+                input_ids, hidden_states, self.lm_head, forward_batch, aux_hidden_states
             )
         else:
             return hidden_states
@@ -705,5 +710,20 @@ class Qwen2MoeForCausalLM(nn.Module):
             num_groups=None,
         )
 
+    def set_eagle3_layers_to_capture(self, layer_ids: Optional[List[int]] = None):
+        if not self.pp_group.is_last_rank:
+            return
+
+        self.capture_aux_hidden_states = True
+        if layer_ids is None:
+            num_layers = self.config.num_hidden_layers
+            self.model.layers_to_capture = [
+                2,
+                num_layers // 2,
+                num_layers - 3,
+            ]  # Specific layers for EAGLE3 support
+        else:
+            self.model.layers_to_capture = [val + 1 for val in layer_ids]
+
 
 EntryClass = Qwen2MoeForCausalLM