[XPU] add gptq(int4) support (#37844)

Signed-off-by: Kunshang Ji <[email protected]>

.buildkite/intel_jobs/test-intel.yaml

@@ -39,7 +39,9 @@ steps:
         python3 examples/basic/offline_inference/generate.py --model nvidia/Llama-3.1-8B-Instruct-FP8 --block-size 64 --enforce-eager --quantization modelopt --kv-cache-dtype fp8 --attention-backend TRITON_ATTN --max-model-len 4096 &&
         python3 examples/basic/offline_inference/generate.py --model superjob/Qwen3-4B-Instruct-2507-GPTQ-Int4 --block-size 64 --enforce-eager --max-model-len 8192 &&
         python3 examples/basic/offline_inference/generate.py --model ibm-research/PowerMoE-3b --block-size 64 --enforce-eager -tp 2 &&
-        python3 examples/basic/offline_inference/generate.py --model ibm-research/PowerMoE-3b --block-size 64 --enforce-eager -tp 2 --enable-expert-parallel'
+        python3 examples/basic/offline_inference/generate.py --model ibm-research/PowerMoE-3b --block-size 64 --enforce-eager -tp 2 --enable-expert-parallel &&
+        python3 examples/basic/offline_inference/generate.py --model superjob/Qwen3-4B-Instruct-2507-GPTQ-Int4 --max-model-len 8192
+        '
   - label: "XPU V1 test"
     depends_on:
       - image-build-xpu

vllm/model_executor/kernels/linear/mixed_precision/MPLinearKernel.py

@@ -50,8 +50,8 @@ class MPLinearKernel(ABC):
             assert w_zp_param_name is not None
         if c.has_g_idx:
             assert w_gidx_param_name is not None
-        self.w_zp_name = w_zp_param_name
-        self.w_gidx_name = w_gidx_param_name
+        self.w_zp_name: str | None = w_zp_param_name
+        self.w_gidx_name: str | None = w_gidx_param_name
 
     @abstractmethod
     def process_weights_after_loading(self, layer: torch.nn.Module) -> None:

vllm/model_executor/kernels/linear/mixed_precision/xpu.py

@@ -61,17 +61,53 @@ class XPUwNa16LinearKernel(MPLinearKernel):
         return True, None
 
     def process_weights_after_loading(self, layer: torch.nn.Module):
-        layer.weight_scale.data = layer.weight_scale.t().contiguous()
+        # Default names since marlin requires empty parameters for these,
+        # TODO: remove this requirement from marlin (allow optional tensors)
+        if self.w_gidx_name is None:
+            self.w_gidx_name = "g_idx"
+        if self.w_zp_name is None:
+            self.w_zp_name = "w_zp"
+
+        need_transpose = False
+        qweight_shape = getattr(layer, self.w_q_name).shape
+        scale_shape = getattr(layer, self.w_s_name).shape
+        # gptq marlin and compressed tensors wna16 expect different default
+        # layouts for weight and scale, so we check the shapes to determine
+        # if we need to transpose
+        if qweight_shape[0] != scale_shape[0]:
+            need_transpose = True
+
+        if need_transpose:
+            getattr(layer, self.w_q_name).data = (
+                getattr(layer, self.w_q_name).data.t().contiguous()
+            )
+            getattr(layer, self.w_s_name).data = getattr(layer, self.w_s_name).data
+        else:
+            getattr(layer, self.w_s_name).data = (
+                getattr(layer, self.w_s_name).data.t().contiguous()
+            )
 
         if self.config.zero_points:
-            layer.weight_zero_point.data = layer.weight_zero_point.t().contiguous()
+            # (FIXME): maybe zero points should also be transposed.
+            getattr(layer, self.w_zp_name).data = (
+                getattr(layer, self.w_zp_name).data.t().contiguous()
+            )
         else:
             weight_zero_point = torch.Tensor([8]).to(torch.int8).to("xpu")
-            layer.weight_zero_point = Parameter(weight_zero_point, requires_grad=False)
+            setattr(
+                layer, self.w_zp_name, Parameter(weight_zero_point, requires_grad=False)
+            )
         if self.config.has_g_idx:
-            layer.g_idx.data = layer.g_idx.t().contiguous()
+            setattr(
+                layer,
+                self.w_gidx_name,
+                Parameter(
+                    getattr(layer, self.w_gidx_name).data.t().contiguous(),
+                    requires_grad=False,
+                ),
+            )
         else:
-            layer.g_idx = None
+            setattr(layer, self.w_gidx_name, None)
 
     def apply_weights(
         self,
@@ -80,14 +116,15 @@ class XPUwNa16LinearKernel(MPLinearKernel):
         bias: torch.Tensor | None = None,
     ) -> torch.Tensor:
         reshaped_x = x.reshape(-1, x.shape[-1])
+        w_q, w_s, w_zp, w_gidx = self._get_weight_params(layer)
         out = torch.ops._xpu_C.int4_gemm_w4a16(
             reshaped_x,
-            layer.weight_packed.t(),
-            bias,
-            layer.weight_scale,
-            layer.weight_zero_point,
+            w_q.t(),
+            bias if bias is not None else None,
+            w_s,
+            w_zp,
             self.config.group_size,
-            layer.g_idx,
+            w_gidx,
         )
         return out
 

vllm/model_executor/layers/quantization/utils/marlin_utils.py

@@ -47,6 +47,9 @@ def query_marlin_supported_quant_types(
     if current_platform.is_cpu():
         return _query_cpu_marlin_supported_quant_types(has_zp, include_fp_type)
 
+    if current_platform.is_xpu():
+        return [scalar_types.uint4, scalar_types.uint4b8]
+
     if not current_platform.is_rocm():
         if device_capability is None:
             capability_tuple = current_platform.get_device_capability()