Duplicate from kernels-community/quantization-eetq

Co-authored-by: Daniël de Kok <danieldk@users.noreply.huggingface.co>

.gitattributes

@@ -0,0 +1,36 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tar filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text
+*.so filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,31 @@
+---
+library_name: kernels
+license: apache-2.0
+---
+
+> [!CAUTION]
+> Starting from September 13, 2026, we will be removing the "model" type repositories of kernels (e.g., kernels-community/flash-attn3). Make sure you're using a latest version of kernels. If you face any disruption, please report them here: https://github.com/huggingface/kernels/issues/new.
+
+This is the repository card of kernels-community/quantization-eetq that has been pushed on the Hub. It was built to be used with the [`kernels` library](https://github.com/huggingface/kernels). This card was automatically generated.
+
+## How to use
+
+```python
+# make sure `kernels` is installed: `pip install -U kernels`
+from kernels import get_kernel
+
+kernel_module = get_kernel("kernels-community/quantization-eetq")
+w8_a16_gemm = kernel_module.w8_a16_gemm
+
+w8_a16_gemm(...)
+```
+
+## Available functions
+- `w8_a16_gemm`
+- `w8_a16_gemm_`
+- `preprocess_weights`
+- `quant_weights`
+
+## Benchmarks
+
+No benchmark available yet.

build/torch210-cxx11-cu126-aarch64-linux/__init__.py

@@ -0,0 +1,3 @@
+from .custom_ops import w8_a16_gemm, w8_a16_gemm_, preprocess_weights, quant_weights
+
+__all__ = ["w8_a16_gemm", "w8_a16_gemm_", "preprocess_weights", "quant_weights"]

build/torch210-cxx11-cu126-aarch64-linux/_ops.py

@@ -0,0 +1,9 @@
+import torch
+from . import _quantization_eetq_cuda_2d6119c
+ops = torch.ops._quantization_eetq_cuda_2d6119c
+
+def add_op_namespace_prefix(op_name: str):
+    """
+    Prefix op by namespace.
+    """
+    return f"_quantization_eetq_cuda_2d6119c::{op_name}"

build/torch210-cxx11-cu126-aarch64-linux/_quantization_eetq_cuda_2d6119c.abi3.so

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:40d168ca6634a23dded922ae1bd52d8e1a2ad30701de31e7e54c826979cf5512
+size 39010048

build/torch210-cxx11-cu126-aarch64-linux/custom_ops.py

@@ -0,0 +1,36 @@
+from typing import List
+import torch
+
+from ._ops import ops
+
+
+def w8_a16_gemm(
+    input: torch.Tensor, weight: torch.Tensor, scale: torch.Tensor
+) -> torch.Tensor:
+    return ops.w8_a16_gemm(input, weight, scale)
+
+
+def w8_a16_gemm_(
+    input: torch.Tensor,
+    weight: torch.Tensor,
+    scale: torch.Tensor,
+    output: torch.Tensor,
+    m: int,
+    n: int,
+    k: int,
+) -> torch.Tensor:
+    return ops.w8_a16_gemm_(input, weight, scale, output, m, n, k)
+
+
+def preprocess_weights(origin_weight: torch.Tensor, is_int4: bool) -> torch.Tensor:
+    return ops.preprocess_weights(origin_weight, is_int4)
+
+
+def quant_weights(
+    origin_weight: torch.Tensor,
+    quant_type: torch.dtype,
+    return_unprocessed_quantized_tensor: bool,
+) -> List[torch.Tensor]:
+    return ops.quant_weights(
+        origin_weight, quant_type, return_unprocessed_quantized_tensor
+    )

build/torch210-cxx11-cu126-aarch64-linux/metadata.json

@@ -0,0 +1,20 @@
+{
+  "name": "quantization-eetq",
+  "id": "_quantization_eetq_cuda_2d6119c",
+  "version": 1,
+  "license": "Apache-2.0",
+  "python-depends": [],
+  "backend": {
+    "type": "cuda",
+    "archs": [
+      "7.0",
+      "7.2",
+      "7.5",
+      "8.0",
+      "8.6",
+      "8.7",
+      "8.9",
+      "9.0+PTX"
+    ]
+  }
+}

build/torch210-cxx11-cu126-aarch64-linux/quantization_eetq/__init__.py

@@ -0,0 +1,26 @@
+import ctypes
+import importlib.util
+import sys
+from pathlib import Path
+from types import ModuleType
+
+
+def _import_from_path(file_path: Path) -> ModuleType:
+    # We cannot use the module name as-is, after adding it to `sys.modules`,
+    # it would also be used for other imports. So, we make a module name that
+    # depends on the path for it to be unique using the hex-encoded hash of
+    # the path.
+    path_hash = "{:x}".format(ctypes.c_size_t(hash(file_path.absolute())).value)
+    module_name = path_hash
+    spec = importlib.util.spec_from_file_location(module_name, file_path)
+    if spec is None:
+        raise ImportError(f"Cannot load spec for {module_name} from {file_path}")
+    module = importlib.util.module_from_spec(spec)
+    if module is None:
+        raise ImportError(f"Cannot load module {module_name} from spec")
+    sys.modules[module_name] = module
+    spec.loader.exec_module(module)  # type: ignore
+    return module
+
+
+globals().update(vars(_import_from_path(Path(__file__).parent.parent / "__init__.py")))

build/torch210-cxx11-cu126-x86_64-linux/__init__.py

@@ -0,0 +1,3 @@
+from .custom_ops import w8_a16_gemm, w8_a16_gemm_, preprocess_weights, quant_weights
+
+__all__ = ["w8_a16_gemm", "w8_a16_gemm_", "preprocess_weights", "quant_weights"]

build/torch210-cxx11-cu126-x86_64-linux/_ops.py

@@ -0,0 +1,9 @@
+import torch
+from . import _quantization_eetq_cuda_2d6119c
+ops = torch.ops._quantization_eetq_cuda_2d6119c
+
+def add_op_namespace_prefix(op_name: str):
+    """
+    Prefix op by namespace.
+    """
+    return f"_quantization_eetq_cuda_2d6119c::{op_name}"

build/torch210-cxx11-cu126-x86_64-linux/_quantization_eetq_cuda_2d6119c.abi3.so

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:755a97594dabf24ea156559e167e41bbdb695fa39c95033efa3a07933c794004
+size 39060416

build/torch210-cxx11-cu126-x86_64-linux/custom_ops.py

@@ -0,0 +1,36 @@
+from typing import List
+import torch
+
+from ._ops import ops
+
+
+def w8_a16_gemm(
+    input: torch.Tensor, weight: torch.Tensor, scale: torch.Tensor
+) -> torch.Tensor:
+    return ops.w8_a16_gemm(input, weight, scale)
+
+
+def w8_a16_gemm_(
+    input: torch.Tensor,
+    weight: torch.Tensor,
+    scale: torch.Tensor,
+    output: torch.Tensor,
+    m: int,
+    n: int,
+    k: int,
+) -> torch.Tensor:
+    return ops.w8_a16_gemm_(input, weight, scale, output, m, n, k)
+
+
+def preprocess_weights(origin_weight: torch.Tensor, is_int4: bool) -> torch.Tensor:
+    return ops.preprocess_weights(origin_weight, is_int4)
+
+
+def quant_weights(
+    origin_weight: torch.Tensor,
+    quant_type: torch.dtype,
+    return_unprocessed_quantized_tensor: bool,
+) -> List[torch.Tensor]:
+    return ops.quant_weights(
+        origin_weight, quant_type, return_unprocessed_quantized_tensor
+    )

build/torch210-cxx11-cu126-x86_64-linux/metadata.json

@@ -0,0 +1,20 @@
+{
+  "name": "quantization-eetq",
+  "id": "_quantization_eetq_cuda_2d6119c",
+  "version": 1,
+  "license": "Apache-2.0",
+  "python-depends": [],
+  "backend": {
+    "type": "cuda",
+    "archs": [
+      "7.0",
+      "7.2",
+      "7.5",
+      "8.0",
+      "8.6",
+      "8.7",
+      "8.9",
+      "9.0+PTX"
+    ]
+  }
+}

build/torch210-cxx11-cu126-x86_64-linux/quantization_eetq/__init__.py

@@ -0,0 +1,26 @@
+import ctypes
+import importlib.util
+import sys
+from pathlib import Path
+from types import ModuleType
+
+
+def _import_from_path(file_path: Path) -> ModuleType:
+    # We cannot use the module name as-is, after adding it to `sys.modules`,
+    # it would also be used for other imports. So, we make a module name that
+    # depends on the path for it to be unique using the hex-encoded hash of
+    # the path.
+    path_hash = "{:x}".format(ctypes.c_size_t(hash(file_path.absolute())).value)
+    module_name = path_hash
+    spec = importlib.util.spec_from_file_location(module_name, file_path)
+    if spec is None:
+        raise ImportError(f"Cannot load spec for {module_name} from {file_path}")
+    module = importlib.util.module_from_spec(spec)
+    if module is None:
+        raise ImportError(f"Cannot load module {module_name} from spec")
+    sys.modules[module_name] = module
+    spec.loader.exec_module(module)  # type: ignore
+    return module
+
+
+globals().update(vars(_import_from_path(Path(__file__).parent.parent / "__init__.py")))

build/torch210-cxx11-cu128-aarch64-linux/__init__.py

@@ -0,0 +1,3 @@
+from .custom_ops import w8_a16_gemm, w8_a16_gemm_, preprocess_weights, quant_weights
+
+__all__ = ["w8_a16_gemm", "w8_a16_gemm_", "preprocess_weights", "quant_weights"]

build/torch210-cxx11-cu128-aarch64-linux/_ops.py

@@ -0,0 +1,9 @@
+import torch
+from . import _quantization_eetq_cuda_2d6119c
+ops = torch.ops._quantization_eetq_cuda_2d6119c
+
+def add_op_namespace_prefix(op_name: str):
+    """
+    Prefix op by namespace.
+    """
+    return f"_quantization_eetq_cuda_2d6119c::{op_name}"

build/torch210-cxx11-cu128-aarch64-linux/_quantization_eetq_cuda_2d6119c.abi3.so

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4f185271e38f588a714e8b360fc7491df3a9592c330c58e805879767f05f92d0
+size 45366048

build/torch210-cxx11-cu128-aarch64-linux/custom_ops.py

@@ -0,0 +1,36 @@
+from typing import List
+import torch
+
+from ._ops import ops
+
+
+def w8_a16_gemm(
+    input: torch.Tensor, weight: torch.Tensor, scale: torch.Tensor
+) -> torch.Tensor:
+    return ops.w8_a16_gemm(input, weight, scale)
+
+
+def w8_a16_gemm_(
+    input: torch.Tensor,
+    weight: torch.Tensor,
+    scale: torch.Tensor,
+    output: torch.Tensor,
+    m: int,
+    n: int,
+    k: int,
+) -> torch.Tensor:
+    return ops.w8_a16_gemm_(input, weight, scale, output, m, n, k)
+
+
+def preprocess_weights(origin_weight: torch.Tensor, is_int4: bool) -> torch.Tensor:
+    return ops.preprocess_weights(origin_weight, is_int4)
+
+
+def quant_weights(
+    origin_weight: torch.Tensor,
+    quant_type: torch.dtype,
+    return_unprocessed_quantized_tensor: bool,
+) -> List[torch.Tensor]:
+    return ops.quant_weights(
+        origin_weight, quant_type, return_unprocessed_quantized_tensor
+    )

build/torch210-cxx11-cu128-aarch64-linux/metadata.json

@@ -0,0 +1,23 @@
+{
+  "name": "quantization-eetq",
+  "id": "_quantization_eetq_cuda_2d6119c",
+  "version": 1,
+  "license": "Apache-2.0",
+  "python-depends": [],
+  "backend": {
+    "type": "cuda",
+    "archs": [
+      "10.0",
+      "10.1",
+      "12.0+PTX",
+      "7.0",
+      "7.2",
+      "7.5",
+      "8.0",
+      "8.6",
+      "8.7",
+      "8.9",
+      "9.0"
+    ]
+  }
+}

build/torch210-cxx11-cu128-aarch64-linux/quantization_eetq/__init__.py

@@ -0,0 +1,26 @@
+import ctypes
+import importlib.util
+import sys
+from pathlib import Path
+from types import ModuleType
+
+
+def _import_from_path(file_path: Path) -> ModuleType:
+    # We cannot use the module name as-is, after adding it to `sys.modules`,
+    # it would also be used for other imports. So, we make a module name that
+    # depends on the path for it to be unique using the hex-encoded hash of
+    # the path.
+    path_hash = "{:x}".format(ctypes.c_size_t(hash(file_path.absolute())).value)
+    module_name = path_hash
+    spec = importlib.util.spec_from_file_location(module_name, file_path)
+    if spec is None:
+        raise ImportError(f"Cannot load spec for {module_name} from {file_path}")
+    module = importlib.util.module_from_spec(spec)
+    if module is None:
+        raise ImportError(f"Cannot load module {module_name} from spec")
+    sys.modules[module_name] = module
+    spec.loader.exec_module(module)  # type: ignore
+    return module
+
+
+globals().update(vars(_import_from_path(Path(__file__).parent.parent / "__init__.py")))

build/torch210-cxx11-cu128-x86_64-linux/__init__.py

@@ -0,0 +1,3 @@
+from .custom_ops import w8_a16_gemm, w8_a16_gemm_, preprocess_weights, quant_weights
+
+__all__ = ["w8_a16_gemm", "w8_a16_gemm_", "preprocess_weights", "quant_weights"]

build/torch210-cxx11-cu128-x86_64-linux/_ops.py

@@ -0,0 +1,9 @@
+import torch
+from . import _quantization_eetq_cuda_2d6119c
+ops = torch.ops._quantization_eetq_cuda_2d6119c
+
+def add_op_namespace_prefix(op_name: str):
+    """
+    Prefix op by namespace.
+    """
+    return f"_quantization_eetq_cuda_2d6119c::{op_name}"

build/torch210-cxx11-cu128-x86_64-linux/_quantization_eetq_cuda_2d6119c.abi3.so

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3691b28fa032d9371e7f91db9b61c7b67926944b6b3a621b69e097a32415c49e
+size 45394000

build/torch210-cxx11-cu128-x86_64-linux/custom_ops.py

@@ -0,0 +1,36 @@
+from typing import List
+import torch
+
+from ._ops import ops
+
+
+def w8_a16_gemm(
+    input: torch.Tensor, weight: torch.Tensor, scale: torch.Tensor
+) -> torch.Tensor:
+    return ops.w8_a16_gemm(input, weight, scale)
+
+
+def w8_a16_gemm_(
+    input: torch.Tensor,
+    weight: torch.Tensor,
+    scale: torch.Tensor,
+    output: torch.Tensor,
+    m: int,
+    n: int,
+    k: int,
+) -> torch.Tensor:
+    return ops.w8_a16_gemm_(input, weight, scale, output, m, n, k)
+
+
+def preprocess_weights(origin_weight: torch.Tensor, is_int4: bool) -> torch.Tensor:
+    return ops.preprocess_weights(origin_weight, is_int4)
+
+
+def quant_weights(
+    origin_weight: torch.Tensor,
+    quant_type: torch.dtype,
+    return_unprocessed_quantized_tensor: bool,
+) -> List[torch.Tensor]:
+    return ops.quant_weights(
+        origin_weight, quant_type, return_unprocessed_quantized_tensor
+    )

build/torch210-cxx11-cu128-x86_64-linux/metadata.json

@@ -0,0 +1,23 @@
+{
+  "name": "quantization-eetq",
+  "id": "_quantization_eetq_cuda_2d6119c",
+  "version": 1,
+  "license": "Apache-2.0",
+  "python-depends": [],
+  "backend": {
+    "type": "cuda",
+    "archs": [
+      "10.0",
+      "10.1",
+      "12.0+PTX",
+      "7.0",
+      "7.2",
+      "7.5",
+      "8.0",
+      "8.6",
+      "8.7",
+      "8.9",
+      "9.0"
+    ]
+  }
+}

build/torch210-cxx11-cu128-x86_64-linux/quantization_eetq/__init__.py

@@ -0,0 +1,26 @@
+import ctypes
+import importlib.util
+import sys
+from pathlib import Path
+from types import ModuleType
+
+
+def _import_from_path(file_path: Path) -> ModuleType:
+    # We cannot use the module name as-is, after adding it to `sys.modules`,
+    # it would also be used for other imports. So, we make a module name that
+    # depends on the path for it to be unique using the hex-encoded hash of
+    # the path.
+    path_hash = "{:x}".format(ctypes.c_size_t(hash(file_path.absolute())).value)
+    module_name = path_hash
+    spec = importlib.util.spec_from_file_location(module_name, file_path)
+    if spec is None:
+        raise ImportError(f"Cannot load spec for {module_name} from {file_path}")
+    module = importlib.util.module_from_spec(spec)
+    if module is None:
+        raise ImportError(f"Cannot load module {module_name} from spec")
+    sys.modules[module_name] = module
+    spec.loader.exec_module(module)  # type: ignore
+    return module
+
+
+globals().update(vars(_import_from_path(Path(__file__).parent.parent / "__init__.py")))

build/torch211-cxx11-cu126-aarch64-linux/__init__.py

@@ -0,0 +1,3 @@
+from .custom_ops import w8_a16_gemm, w8_a16_gemm_, preprocess_weights, quant_weights
+
+__all__ = ["w8_a16_gemm", "w8_a16_gemm_", "preprocess_weights", "quant_weights"]

build/torch211-cxx11-cu126-aarch64-linux/_ops.py

@@ -0,0 +1,9 @@
+import torch
+from . import _quantization_eetq_cuda_2d6119c
+ops = torch.ops._quantization_eetq_cuda_2d6119c
+
+def add_op_namespace_prefix(op_name: str):
+    """
+    Prefix op by namespace.
+    """
+    return f"_quantization_eetq_cuda_2d6119c::{op_name}"

build/torch211-cxx11-cu126-aarch64-linux/_quantization_eetq_cuda_2d6119c.abi3.so

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9166428182ff8357a8124b5fa9b45557867419cd7095c16264c0d7b4a8b1985c
+size 39006256

build/torch211-cxx11-cu126-aarch64-linux/custom_ops.py

@@ -0,0 +1,36 @@
+from typing import List
+import torch
+
+from ._ops import ops
+
+
+def w8_a16_gemm(
+    input: torch.Tensor, weight: torch.Tensor, scale: torch.Tensor
+) -> torch.Tensor:
+    return ops.w8_a16_gemm(input, weight, scale)
+
+
+def w8_a16_gemm_(
+    input: torch.Tensor,
+    weight: torch.Tensor,
+    scale: torch.Tensor,
+    output: torch.Tensor,
+    m: int,
+    n: int,
+    k: int,
+) -> torch.Tensor:
+    return ops.w8_a16_gemm_(input, weight, scale, output, m, n, k)
+
+
+def preprocess_weights(origin_weight: torch.Tensor, is_int4: bool) -> torch.Tensor:
+    return ops.preprocess_weights(origin_weight, is_int4)
+
+
+def quant_weights(
+    origin_weight: torch.Tensor,
+    quant_type: torch.dtype,
+    return_unprocessed_quantized_tensor: bool,
+) -> List[torch.Tensor]:
+    return ops.quant_weights(
+        origin_weight, quant_type, return_unprocessed_quantized_tensor
+    )

build/torch211-cxx11-cu126-aarch64-linux/metadata.json

@@ -0,0 +1,20 @@
+{
+  "name": "quantization-eetq",
+  "id": "_quantization_eetq_cuda_2d6119c",
+  "version": 1,
+  "license": "Apache-2.0",
+  "python-depends": [],
+  "backend": {
+    "type": "cuda",
+    "archs": [
+      "7.0",
+      "7.2",
+      "7.5",
+      "8.0",
+      "8.6",
+      "8.7",
+      "8.9",
+      "9.0+PTX"
+    ]
+  }
+}

build/torch211-cxx11-cu126-aarch64-linux/quantization_eetq/__init__.py

@@ -0,0 +1,26 @@
+import ctypes
+import importlib.util
+import sys
+from pathlib import Path
+from types import ModuleType
+
+
+def _import_from_path(file_path: Path) -> ModuleType:
+    # We cannot use the module name as-is, after adding it to `sys.modules`,
+    # it would also be used for other imports. So, we make a module name that
+    # depends on the path for it to be unique using the hex-encoded hash of
+    # the path.
+    path_hash = "{:x}".format(ctypes.c_size_t(hash(file_path.absolute())).value)
+    module_name = path_hash
+    spec = importlib.util.spec_from_file_location(module_name, file_path)
+    if spec is None:
+        raise ImportError(f"Cannot load spec for {module_name} from {file_path}")
+    module = importlib.util.module_from_spec(spec)
+    if module is None:
+        raise ImportError(f"Cannot load module {module_name} from spec")
+    sys.modules[module_name] = module
+    spec.loader.exec_module(module)  # type: ignore
+    return module
+
+
+globals().update(vars(_import_from_path(Path(__file__).parent.parent / "__init__.py")))

build/torch211-cxx11-cu126-x86_64-linux/__init__.py

@@ -0,0 +1,3 @@
+from .custom_ops import w8_a16_gemm, w8_a16_gemm_, preprocess_weights, quant_weights
+
+__all__ = ["w8_a16_gemm", "w8_a16_gemm_", "preprocess_weights", "quant_weights"]

build/torch211-cxx11-cu126-x86_64-linux/_ops.py

@@ -0,0 +1,9 @@
+import torch
+from . import _quantization_eetq_cuda_2d6119c
+ops = torch.ops._quantization_eetq_cuda_2d6119c
+
+def add_op_namespace_prefix(op_name: str):
+    """
+    Prefix op by namespace.
+    """
+    return f"_quantization_eetq_cuda_2d6119c::{op_name}"

build/torch211-cxx11-cu126-x86_64-linux/_quantization_eetq_cuda_2d6119c.abi3.so

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:68709698dac1b4037fab603423fa85b4f9b765dc05162173ec7681be660e672f
+size 39053336

build/torch211-cxx11-cu126-x86_64-linux/custom_ops.py

@@ -0,0 +1,36 @@
+from typing import List
+import torch
+
+from ._ops import ops
+
+
+def w8_a16_gemm(
+    input: torch.Tensor, weight: torch.Tensor, scale: torch.Tensor
+) -> torch.Tensor:
+    return ops.w8_a16_gemm(input, weight, scale)
+
+
+def w8_a16_gemm_(
+    input: torch.Tensor,
+    weight: torch.Tensor,
+    scale: torch.Tensor,
+    output: torch.Tensor,
+    m: int,
+    n: int,
+    k: int,
+) -> torch.Tensor:
+    return ops.w8_a16_gemm_(input, weight, scale, output, m, n, k)
+
+
+def preprocess_weights(origin_weight: torch.Tensor, is_int4: bool) -> torch.Tensor:
+    return ops.preprocess_weights(origin_weight, is_int4)
+
+
+def quant_weights(
+    origin_weight: torch.Tensor,
+    quant_type: torch.dtype,
+    return_unprocessed_quantized_tensor: bool,
+) -> List[torch.Tensor]:
+    return ops.quant_weights(
+        origin_weight, quant_type, return_unprocessed_quantized_tensor
+    )

build/torch211-cxx11-cu126-x86_64-linux/metadata.json

@@ -0,0 +1,20 @@
+{
+  "name": "quantization-eetq",
+  "id": "_quantization_eetq_cuda_2d6119c",
+  "version": 1,
+  "license": "Apache-2.0",
+  "python-depends": [],
+  "backend": {
+    "type": "cuda",
+    "archs": [
+      "7.0",
+      "7.2",
+      "7.5",
+      "8.0",
+      "8.6",
+      "8.7",
+      "8.9",
+      "9.0+PTX"
+    ]
+  }
+}

build/torch211-cxx11-cu126-x86_64-linux/quantization_eetq/__init__.py

@@ -0,0 +1,26 @@
+import ctypes
+import importlib.util
+import sys
+from pathlib import Path
+from types import ModuleType
+
+
+def _import_from_path(file_path: Path) -> ModuleType:
+    # We cannot use the module name as-is, after adding it to `sys.modules`,
+    # it would also be used for other imports. So, we make a module name that
+    # depends on the path for it to be unique using the hex-encoded hash of
+    # the path.
+    path_hash = "{:x}".format(ctypes.c_size_t(hash(file_path.absolute())).value)
+    module_name = path_hash
+    spec = importlib.util.spec_from_file_location(module_name, file_path)
+    if spec is None:
+        raise ImportError(f"Cannot load spec for {module_name} from {file_path}")
+    module = importlib.util.module_from_spec(spec)
+    if module is None:
+        raise ImportError(f"Cannot load module {module_name} from spec")
+    sys.modules[module_name] = module
+    spec.loader.exec_module(module)  # type: ignore
+    return module
+
+
+globals().update(vars(_import_from_path(Path(__file__).parent.parent / "__init__.py")))

build/torch211-cxx11-cu128-aarch64-linux/__init__.py

@@ -0,0 +1,3 @@
+from .custom_ops import w8_a16_gemm, w8_a16_gemm_, preprocess_weights, quant_weights
+
+__all__ = ["w8_a16_gemm", "w8_a16_gemm_", "preprocess_weights", "quant_weights"]

build/torch211-cxx11-cu128-aarch64-linux/_ops.py

@@ -0,0 +1,9 @@
+import torch
+from . import _quantization_eetq_cuda_2d6119c
+ops = torch.ops._quantization_eetq_cuda_2d6119c
+
+def add_op_namespace_prefix(op_name: str):
+    """
+    Prefix op by namespace.
+    """
+    return f"_quantization_eetq_cuda_2d6119c::{op_name}"

build/torch211-cxx11-cu128-aarch64-linux/_quantization_eetq_cuda_2d6119c.abi3.so

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e1eb5bfd99bc0c2ad9f430b85f50d5e5075103bd63f0561bf2678472d7101e97
+size 45296712

build/torch211-cxx11-cu128-aarch64-linux/custom_ops.py

@@ -0,0 +1,36 @@
+from typing import List
+import torch
+
+from ._ops import ops
+
+
+def w8_a16_gemm(
+    input: torch.Tensor, weight: torch.Tensor, scale: torch.Tensor
+) -> torch.Tensor:
+    return ops.w8_a16_gemm(input, weight, scale)
+
+
+def w8_a16_gemm_(
+    input: torch.Tensor,
+    weight: torch.Tensor,
+    scale: torch.Tensor,
+    output: torch.Tensor,
+    m: int,
+    n: int,
+    k: int,
+) -> torch.Tensor:
+    return ops.w8_a16_gemm_(input, weight, scale, output, m, n, k)
+
+
+def preprocess_weights(origin_weight: torch.Tensor, is_int4: bool) -> torch.Tensor:
+    return ops.preprocess_weights(origin_weight, is_int4)
+
+
+def quant_weights(
+    origin_weight: torch.Tensor,
+    quant_type: torch.dtype,
+    return_unprocessed_quantized_tensor: bool,
+) -> List[torch.Tensor]:
+    return ops.quant_weights(
+        origin_weight, quant_type, return_unprocessed_quantized_tensor
+    )

build/torch211-cxx11-cu128-aarch64-linux/metadata.json

@@ -0,0 +1,23 @@
+{
+  "name": "quantization-eetq",
+  "id": "_quantization_eetq_cuda_2d6119c",
+  "version": 1,
+  "license": "Apache-2.0",
+  "python-depends": [],
+  "backend": {
+    "type": "cuda",
+    "archs": [
+      "10.0",
+      "10.1",
+      "12.0+PTX",
+      "7.0",
+      "7.2",
+      "7.5",
+      "8.0",
+      "8.6",
+      "8.7",
+      "8.9",
+      "9.0"
+    ]
+  }
+}

build/torch211-cxx11-cu128-aarch64-linux/quantization_eetq/__init__.py

@@ -0,0 +1,26 @@
+import ctypes
+import importlib.util
+import sys
+from pathlib import Path
+from types import ModuleType
+
+
+def _import_from_path(file_path: Path) -> ModuleType:
+    # We cannot use the module name as-is, after adding it to `sys.modules`,
+    # it would also be used for other imports. So, we make a module name that
+    # depends on the path for it to be unique using the hex-encoded hash of
+    # the path.
+    path_hash = "{:x}".format(ctypes.c_size_t(hash(file_path.absolute())).value)
+    module_name = path_hash
+    spec = importlib.util.spec_from_file_location(module_name, file_path)
+    if spec is None:
+        raise ImportError(f"Cannot load spec for {module_name} from {file_path}")
+    module = importlib.util.module_from_spec(spec)
+    if module is None:
+        raise ImportError(f"Cannot load module {module_name} from spec")
+    sys.modules[module_name] = module
+    spec.loader.exec_module(module)  # type: ignore
+    return module
+
+
+globals().update(vars(_import_from_path(Path(__file__).parent.parent / "__init__.py")))

build/torch211-cxx11-cu128-x86_64-linux/__init__.py

@@ -0,0 +1,3 @@
+from .custom_ops import w8_a16_gemm, w8_a16_gemm_, preprocess_weights, quant_weights
+
+__all__ = ["w8_a16_gemm", "w8_a16_gemm_", "preprocess_weights", "quant_weights"]

build/torch211-cxx11-cu128-x86_64-linux/_ops.py

@@ -0,0 +1,9 @@
+import torch
+from . import _quantization_eetq_cuda_2d6119c
+ops = torch.ops._quantization_eetq_cuda_2d6119c
+
+def add_op_namespace_prefix(op_name: str):
+    """
+    Prefix op by namespace.
+    """
+    return f"_quantization_eetq_cuda_2d6119c::{op_name}"

build/torch211-cxx11-cu128-x86_64-linux/_quantization_eetq_cuda_2d6119c.abi3.so

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2b694cb766402639a53ead4ce5febcf4b278acec249f7dfac8652ade997af22e
+size 45382832

build/torch211-cxx11-cu128-x86_64-linux/custom_ops.py

@@ -0,0 +1,36 @@
+from typing import List
+import torch
+
+from ._ops import ops
+
+
+def w8_a16_gemm(
+    input: torch.Tensor, weight: torch.Tensor, scale: torch.Tensor
+) -> torch.Tensor:
+    return ops.w8_a16_gemm(input, weight, scale)
+
+
+def w8_a16_gemm_(
+    input: torch.Tensor,
+    weight: torch.Tensor,
+    scale: torch.Tensor,
+    output: torch.Tensor,
+    m: int,
+    n: int,
+    k: int,
+) -> torch.Tensor:
+    return ops.w8_a16_gemm_(input, weight, scale, output, m, n, k)
+
+
+def preprocess_weights(origin_weight: torch.Tensor, is_int4: bool) -> torch.Tensor:
+    return ops.preprocess_weights(origin_weight, is_int4)
+
+
+def quant_weights(
+    origin_weight: torch.Tensor,
+    quant_type: torch.dtype,
+    return_unprocessed_quantized_tensor: bool,
+) -> List[torch.Tensor]:
+    return ops.quant_weights(
+        origin_weight, quant_type, return_unprocessed_quantized_tensor
+    )

build/torch211-cxx11-cu128-x86_64-linux/metadata.json

@@ -0,0 +1,23 @@
+{
+  "name": "quantization-eetq",
+  "id": "_quantization_eetq_cuda_2d6119c",
+  "version": 1,
+  "license": "Apache-2.0",
+  "python-depends": [],
+  "backend": {
+    "type": "cuda",
+    "archs": [
+      "10.0",
+      "10.1",
+      "12.0+PTX",
+      "7.0",
+      "7.2",
+      "7.5",
+      "8.0",
+      "8.6",
+      "8.7",
+      "8.9",
+      "9.0"
+    ]
+  }
+}

build/torch211-cxx11-cu128-x86_64-linux/quantization_eetq/__init__.py

@@ -0,0 +1,26 @@
+import ctypes
+import importlib.util
+import sys
+from pathlib import Path
+from types import ModuleType
+
+
+def _import_from_path(file_path: Path) -> ModuleType:
+    # We cannot use the module name as-is, after adding it to `sys.modules`,
+    # it would also be used for other imports. So, we make a module name that
+    # depends on the path for it to be unique using the hex-encoded hash of
+    # the path.
+    path_hash = "{:x}".format(ctypes.c_size_t(hash(file_path.absolute())).value)
+    module_name = path_hash
+    spec = importlib.util.spec_from_file_location(module_name, file_path)
+    if spec is None:
+        raise ImportError(f"Cannot load spec for {module_name} from {file_path}")
+    module = importlib.util.module_from_spec(spec)
+    if module is None:
+        raise ImportError(f"Cannot load module {module_name} from spec")
+    sys.modules[module_name] = module
+    spec.loader.exec_module(module)  # type: ignore
+    return module
+
+
+globals().update(vars(_import_from_path(Path(__file__).parent.parent / "__init__.py")))