feat: support if vlm models

server/text_generation_server/layers/lora.py

@@ -41,6 +41,8 @@ def forward_layer_type(
         start_idx: int,
         end_idx: int,
     ) -> torch.Tensor:
+        if adapter_data is None:
+            return result
         data = adapter_data.data.get(layer_type)
         data: Optional["BatchLoraWeights"] = (
             data.get(LORA) if data is not None else None

server/text_generation_server/models/flash_llama.py

@@ -108,7 +108,17 @@ def adapter_target_to_layer(self) -> Dict[str, Tuple[str, torch.Tensor]]:
         layer_weights = {}
 
         prefix = "model.layers"
-        for i, layer in enumerate(self.model.model.layers):
+
+        # This accounts for VLMs (e.g. LlavaNext, Idefics2)
+        # that have a language_model inside of the larger model.
+        if hasattr(self.model, "language_model"):
+            _model = self.model.language_model
+        elif hasattr(self.model, "text_model"):
+            _model = self.model.text_model
+        else:
+            _model = self.model
+
+        for i, layer in enumerate(_model.model.layers):
             layer_weights[(i, "q_proj")] = (
                 f"{prefix}.{i}.self_attn.q_proj",
                 layer.self_attn.query_key_value,
@@ -139,7 +149,7 @@ def adapter_target_to_layer(self) -> Dict[str, Tuple[str, torch.Tensor]]:
                 layer.mlp.down_proj,
             )
 
-        layer_weights[(0, LM_HEAD)] = ("lm_head", self.model.lm_head)
+        layer_weights[(0, "lm_head")] = ("lm_head", _model.lm_head)
         return layer_weights
 
     @property
@@ -151,7 +161,7 @@ def default_traced_adapter_layers(self) -> List[str]:
         return ["q_proj", "v_proj"]
 
     def get_num_layers_for_type(self, layer_type: str) -> int:
-        return 1 if layer_type == LM_HEAD else len(self.model.model.layers)
+        return 1 if layer_type == "lm_head" else len(self.model.model.layers)
 
     def is_row_parallel(self, layer_type: str) -> bool:
         return layer_type in ROW_PARALLEL

server/text_generation_server/models/flash_mistral.py

@@ -119,7 +119,17 @@ def adapter_target_to_layer(self) -> Dict[str, Tuple[str, torch.Tensor]]:
         layer_weights = {}
 
         prefix = "model.layers"
-        for i, layer in enumerate(self.model.model.layers):
+
+        # This accounts for VLMs (e.g. LlavaNext, Idefics2)
+        # that have a language_model inside of the larger model.
+        if hasattr(self.model, "language_model"):
+            _model = self.model.language_model
+        elif hasattr(self.model, "text_model"):
+            _model = self.model.text_model
+        else:
+            _model = self.model
+
+        for i, layer in enumerate(_model.model.layers):
             layer_weights[(i, "q_proj")] = (
                 f"{prefix}.{i}.self_attn.q_proj",
                 layer.self_attn.query_key_value,
@@ -150,7 +160,7 @@ def adapter_target_to_layer(self) -> Dict[str, Tuple[str, torch.Tensor]]:
                 layer.mlp.down_proj,
             )
 
-        layer_weights[(0, "lm_head")] = ("lm_head", self.model.lm_head)
+        layer_weights[(0, "lm_head")] = ("lm_head", _model.lm_head)
         return layer_weights
 
     @property

server/text_generation_server/models/mamba.py

@@ -452,6 +452,7 @@ def __init__(
         model = MambaModel(config, weights)
         torch.distributed.barrier(group=self.process_group)
         super(Mamba, self).__init__(
+            model_id=model_id,
             model=model,
             tokenizer=tokenizer,
             requires_padding=True,

server/text_generation_server/models/vlm_causal_lm.py

@@ -257,7 +257,9 @@ def batch_type(self) -> Type[VlmCausalLMBatch]:
         return VlmCausalLMBatch
 
     def forward(
-        self, batch: VlmCausalLMBatch
+        self,
+        batch: VlmCausalLMBatch,
+        adapter_data: Optional[Dict[str, torch.Tensor]] = None,
     ) -> Tuple[torch.Tensor, Optional[torch.Tensor]]:
         # Model Forward
         if batch.speculative_ids is not None: