remove vllm model_loader deps and support qqq quantization

python/sglang/srt/layers/linear.py

@@ -42,6 +42,7 @@
     "Fp8LinearMethod",
     "MarlinLinearMethod",
     "GPTQLinearMethod",
+    "QQQLinearMethod",
 ]
 
 

python/sglang/srt/layers/quantization/__init__.py

@@ -19,10 +19,10 @@
 from vllm.model_executor.layers.quantization.gptq_marlin import GPTQMarlinConfig
 from vllm.model_executor.layers.quantization.gptq_marlin_24 import GPTQMarlin24Config
 from vllm.model_executor.layers.quantization.marlin import MarlinConfig
-from vllm.model_executor.layers.quantization.qqq import QQQConfig
 from vllm.model_executor.layers.quantization.tpu_int8 import Int8TpuConfig
 
 from sglang.srt.layers.quantization.base_config import QuantizationConfig
+from sglang.srt.layers.quantization.qqq import QQQConfig
 
 QUANTIZATION_METHODS: Dict[str, Type[QuantizationConfig]] = {
     "aqlm": AQLMConfig,

python/sglang/srt/layers/quantization/qqq.py

@@ -0,0 +1,300 @@
+# Adapted from https://github.com/vllm-project/vllm/blob/v0.6.3.post1/vllm/model_executor/layers/quantization/qqq.py
+
+import logging
+from typing import Any, Dict, List, Optional
+
+import torch
+from torch.nn.parameter import Parameter
+from torchao.ops import marlin_qqq_gemm
+from torchao.quantization.utils import dynamically_quantize_per_channel
+from vllm.model_executor.layers.linear import LinearBase, LinearMethodBase
+from vllm.model_executor.parameter import (
+    BasevLLMParameter,
+    ChannelQuantScaleParameter,
+    GroupQuantScaleParameter,
+    PackedvLLMParameter,
+)
+
+from sglang.srt.layers.quantization.base_config import QuantizationConfig
+
+logger = logging.getLogger(__name__)
+
+MARLIN_QQQ_TILE = 16
+MARLIN_QQQ_MIN_THREAD_N = 64
+MARLIN_QQQ_MIN_THREAD_K = 128
+MARLIN_QQQ_MAX_PARALLEL = 16
+
+MARLIN_QQQ_SUPPORTED_NUM_BITS = [4]
+MARLIN_QQQ_SUPPORTED_GROUP_SIZES = [-1, 128]
+MARLIN_QQQ_SUPPORTED_SYM = [True]
+
+
+class QQQConfig(QuantizationConfig):
+    """Config class for QQQ
+
+    Reference: https://arxiv.org/pdf/2406.09904
+    """
+
+    def __init__(
+        self,
+        weight_bits: int,
+        group_size: int,
+        is_sym: bool = True,
+    ) -> None:
+        self.weight_bits = weight_bits
+        self.group_size = group_size
+        self.is_sym = is_sym
+
+        # Verify
+        if self.weight_bits not in MARLIN_QQQ_SUPPORTED_NUM_BITS:
+            raise ValueError(
+                f"QQQ does not support weight_bits = {self.weight_bits}. "
+                f"Only weight_bits = {MARLIN_QQQ_SUPPORTED_NUM_BITS} "
+                "are supported."
+            )
+        if self.group_size not in MARLIN_QQQ_SUPPORTED_GROUP_SIZES:
+            raise ValueError(
+                f"QQQ does not support group_size = {self.group_size}. "
+                f"Only group_sizes = {MARLIN_QQQ_SUPPORTED_GROUP_SIZES} "
+                "are supported."
+            )
+        if self.is_sym not in MARLIN_QQQ_SUPPORTED_SYM:
+            raise ValueError(
+                f"QQQ does not support is_sym = {self.is_sym}. "
+                f"Only sym = {MARLIN_QQQ_SUPPORTED_SYM} are supported."
+            )
+
+        # 4 Bits packed into 32 bit datatype.
+        self.pack_factor = 32 // self.weight_bits
+
+        # Tile size used by QQQ kernels.
+        self.tile_size = MARLIN_QQQ_TILE
+
+        # Min out_features dim
+        self.min_n_threads = MARLIN_QQQ_MIN_THREAD_N
+
+        # Min in_features dim
+        self.min_k_threads = MARLIN_QQQ_MIN_THREAD_K
+
+        # Max parallel problems to solve at once (improves large
+        # batch performance)
+        self.max_parallel = MARLIN_QQQ_MAX_PARALLEL
+
+        # Permutation length used by the QQQ kernels.
+        self.perm_len = 1024
+
+    def __repr__(self) -> str:
+        return "QQQConfig(weight_bits={}, group_size={})".format(
+            self.weight_bits, self.group_size
+        )
+
+    @classmethod
+    def get_name(cls) -> str:
+        return "qqq"
+
+    @classmethod
+    def get_supported_act_dtypes(cls) -> List[torch.dtype]:
+        return [torch.half]
+
+    @classmethod
+    def get_min_capability(cls) -> int:
+        return 80
+
+    @classmethod
+    def get_config_filenames(cls) -> List[str]:
+        """List of filenames to search for in the model directory."""
+        return [
+            "quant_config.json",
+            "quantize_config.json",
+        ]
+
+    @classmethod
+    def from_config(cls, config: Dict[str, Any]) -> "QQQConfig":
+        weight_bits = cls.get_from_keys(config, ["wbits"])
+        group_size = cls.get_from_keys(config, ["group_size"])
+        return cls(weight_bits, group_size)
+
+    def get_quant_method(
+        self, layer: torch.nn.Module, prefix: str
+    ) -> Optional["QQQLinearMethod"]:
+        if isinstance(layer, LinearBase):
+            return QQQLinearMethod(self)
+        return None
+
+    def get_scaled_act_names(self) -> List[str]:
+        return []
+
+
+class QQQLinearMethod(LinearMethodBase):
+    """Linear method for QQQ.
+
+    Args:
+        quant_config: The QQQ quantization config.
+    """
+
+    def __init__(self, quant_config: QQQConfig):
+        self.quant_config = quant_config
+
+    def create_weights(
+        self,
+        layer: torch.nn.Module,
+        input_size_per_partition: int,
+        output_partition_sizes: List[int],
+        input_size: int,
+        output_size: int,
+        params_dtype: torch.dtype,
+        **extra_weight_attrs,
+    ):
+        weight_loader = extra_weight_attrs["weight_loader"]
+        if params_dtype != torch.float16:
+            raise ValueError(
+                f"The params dtype must be float16, but got {params_dtype}"
+            )
+
+        # Validate output_size_per_partition
+        output_size_per_partition = sum(output_partition_sizes)
+        if output_size_per_partition % self.quant_config.min_n_threads != 0:
+            raise ValueError(
+                f"Weight output_size_per_partition = "
+                f"{output_size_per_partition} is not divisible by "
+                f"min_n_threads = {self.quant_config.min_n_threads}."
+            )
+        if output_size_per_partition % self.quant_config.pack_factor != 0:
+            raise ValueError(
+                f"Weight output_size_per_partition = "
+                f"{output_size_per_partition} is not divisible by "
+                f"pack_factor = {self.quant_config.pack_factor}."
+            )
+
+        # Validate input_size_per_partition
+        if input_size_per_partition % self.quant_config.min_k_threads != 0:
+            raise ValueError(
+                f"Weight input_size_per_partition = "
+                f"{input_size_per_partition} is not divisible by "
+                f"min_k_threads = {self.quant_config.min_k_threads}."
+            )
+        if (
+            self.quant_config.group_size != -1
+            and input_size_per_partition % self.quant_config.group_size != 0
+        ):
+            raise ValueError(
+                f"Weight input_size_per_partition = "
+                f"{input_size_per_partition} is not divisible by "
+                f"group_size = {self.quant_config.group_size}."
+            )
+
+        # Check that we have at least 4 tiles horizontally in the shard
+        num_tiles_per_perm = self.quant_config.perm_len // (
+            self.quant_config.tile_size**2
+        )
+        if output_size_per_partition % num_tiles_per_perm != 0:
+            raise ValueError("Each permutation group must reside on the same gpu")
+
+        # Quantized 4Bit weights packed into Int32.
+        qweight = PackedvLLMParameter(
+            data=torch.empty(
+                input_size_per_partition // self.quant_config.tile_size,
+                output_size_per_partition
+                * self.quant_config.tile_size
+                // self.quant_config.pack_factor,
+                device="cuda",
+                dtype=torch.int32,
+            ),
+            input_dim=0,
+            output_dim=1,
+            packed_dim=1,
+            packed_factor=self.quant_config.pack_factor,
+            marlin_tile_size=self.quant_config.tile_size,
+            weight_loader=weight_loader,
+        )
+
+        s_channel = ChannelQuantScaleParameter(
+            data=torch.empty(
+                1,
+                output_size_per_partition,
+                device="cuda",
+                dtype=torch.float,
+            ),
+            weight_loader=weight_loader,
+            output_dim=1,
+        )
+
+        if self.quant_config.group_size == -1:
+            s_group_data = torch.tensor(
+                [],
+                device="cuda",
+                dtype=torch.half,
+            )
+        else:
+            s_group_data = torch.empty(
+                input_size_per_partition // self.quant_config.group_size,
+                output_size_per_partition,
+                device="cuda",
+                dtype=torch.half,
+            )
+
+        s_group_attr = {"data": s_group_data, "weight_loader": weight_loader}
+
+        if self.quant_config.group_size == -1:
+            s_group = BasevLLMParameter(**s_group_attr)
+        else:
+            s_group = GroupQuantScaleParameter(
+                output_dim=1, input_dim=0, **s_group_attr
+            )
+
+        # Allocate workspace (Used for internal locking mechanism)
+        max_workspace_size = (
+            output_size_per_partition // self.quant_config.min_n_threads
+        ) * self.quant_config.max_parallel
+
+        workspace = BasevLLMParameter(
+            data=torch.zeros(max_workspace_size, device="cuda", dtype=torch.int),
+            weight_loader=weight_loader,
+        )
+
+        layer.register_parameter("B", qweight)
+        layer.register_parameter("s_channel", s_channel)
+        layer.register_parameter("s_group", s_group)
+        layer.register_parameter("workspace", workspace)
+
+    def process_weights_after_loading(self, layer: torch.nn.Module) -> None:
+        # required by torch.compile
+        layer.B = Parameter(layer.B.data, requires_grad=False)
+        layer.s_channel = Parameter(layer.s_channel.data, requires_grad=False)
+        layer.s_group = Parameter(layer.s_group.data, requires_grad=False)
+        layer.workspace = Parameter(layer.workspace.data, requires_grad=False)
+
+    def apply(
+        self,
+        layer: torch.nn.Module,
+        x: torch.Tensor,
+        bias: Optional[torch.Tensor] = None,
+    ) -> torch.Tensor:
+        qweight = layer.B
+        s_ch = layer.s_channel
+        s_group = layer.s_group
+        workspace = layer.workspace
+
+        x_2d = x.view(-1, x.shape[-1])
+
+        size_m = x_2d.shape[0]
+        size_k = x_2d.shape[1]
+        size_n = s_ch.shape[1]
+
+        x_int8, s_tok, _ = dynamically_quantize_per_channel(
+            x_2d, quant_min=-127, quant_max=127, target_dtype=torch.int8
+        )
+        # TODO(HandH1998): As the `dynamically_quantize_per_channel` function in torchao doesn't support defining the `scale_dtype`,
+        # we have to convert `s_tok` to `torch.float32`, which is required by `marlin_qqq_gemm`. Remove it when torchao supports defining the `scale_dtype`.
+        s_tok = s_tok.to(torch.float32)
+
+        output_2d = marlin_qqq_gemm(
+            x_int8, qweight, s_tok, s_ch, s_group, workspace, size_m, size_n, size_k
+        )
+
+        output = output_2d.view(x.shape[:-1] + (output_2d.shape[1],))
+
+        if bias is not None:
+            output.add_(bias)  # In-place add
+
+        return output

python/sglang/srt/lora/lora.py

@@ -31,7 +31,6 @@
     ParallelLMHead,
     VocabParallelEmbedding,
 )
-from vllm.model_executor.model_loader.loader import DefaultModelLoader
 
 from sglang.srt.layers.linear import (
     ColumnParallelLinear,
@@ -40,6 +39,7 @@
     RowParallelLinear,
 )
 from sglang.srt.model_executor.forward_batch_info import ForwardBatch, ForwardMode
+from sglang.srt.model_loader.loader import DefaultModelLoader
 
 
 class BaseLayerWithLoRA(nn.Module):