repo stringclasses 454
values | file_path stringlengths 5 201 | extension stringclasses 1
value | content stringlengths 8 509k | num_lines int64 3 16.9k | size_bytes int64 8 511k |
|---|---|---|---|---|---|
sglang | python/sglang/multimodal_gen/runtime/disaggregation/orchestrator.py | .py | # SPDX-License-Identifier: Apache-2.0
"""Central request router for disaggregated diffusion pipelines."""
import json
import logging
import pickle
import threading
import time
from collections import deque
from dataclasses import dataclass
import zmq
from sglang.multimodal_gen.runtime.disaggregation.dispatch_policy ... | 1,077 | 38,559 |
sglang | python/sglang/multimodal_gen/runtime/disaggregation/disagg_args.py | .py | # SPDX-License-Identifier: Apache-2.0
"""Compatibility shim for disaggregated diffusion argument helpers."""
from __future__ import annotations
import argparse
from sglang.multimodal_gen.runtime.disaggregation.roles import RoleType
from sglang.multimodal_gen.runtime.server_args.disagg import DisaggServerArgsMixin
#... | 29 | 1,031 |
sglang | python/sglang/multimodal_gen/runtime/disaggregation/transport/allocator.py | .py | # SPDX-License-Identifier: Apache-2.0
"""Buddy-system memory allocator for TransferTensorBuffer."""
from __future__ import annotations
import logging
import threading
from dataclasses import dataclass
from functools import lru_cache
logger = logging.getLogger(__name__)
@dataclass
class Block:
offset: int # by... | 187 | 6,010 |
sglang | python/sglang/multimodal_gen/runtime/disaggregation/transport/manager.py | .py | # SPDX-License-Identifier: Apache-2.0
"""Per-instance transfer manager for disaggregated diffusion roles."""
import logging
import threading
from dataclasses import dataclass, field
import torch
from sglang.multimodal_gen.runtime.disaggregation.transport.buffer import (
SlotHandle,
TransferTensorBuffer,
)
fr... | 276 | 8,579 |
sglang | python/sglang/multimodal_gen/runtime/disaggregation/transport/__init__.py | .py | # SPDX-License-Identifier: Apache-2.0
"""Transport layer for disaggregated diffusion pipelines."""
| 3 | 99 |
sglang | python/sglang/multimodal_gen/runtime/disaggregation/transport/protocol.py | .py | # SPDX-License-Identifier: Apache-2.0
"""Transfer protocol messages for disaggregated diffusion.
All messages are sent as ZMQ multipart with a b"__transfer__" discriminator
in frame[0] and JSON payload in frame[1].
"""
import json
from dataclasses import asdict, dataclass, field
from typing import Any
TRANSFER_MAGIC... | 140 | 3,767 |
sglang | python/sglang/multimodal_gen/runtime/disaggregation/transport/buffer.py | .py | # SPDX-License-Identifier: Apache-2.0
"""TransferTensorBuffer: memory staging area for disaggregated tensor transfer."""
from __future__ import annotations
import logging
from dataclasses import dataclass, field
import torch
from sglang.multimodal_gen.runtime.disaggregation.transport.allocator import (
BuddyAll... | 261 | 8,528 |
sglang | python/sglang/multimodal_gen/runtime/disaggregation/transport/engine.py | .py | # SPDX-License-Identifier: Apache-2.0
"""Transfer engine abstraction for tensor transfer between role instances."""
import logging
from abc import ABC, abstractmethod
logger = logging.getLogger(__name__)
_MOONCAKE_AVAILABLE = None
def _check_mooncake() -> bool:
global _MOONCAKE_AVAILABLE
if _MOONCAKE_AVAIL... | 127 | 3,523 |
sglang | python/sglang/multimodal_gen/runtime/disaggregation/transport/codec.py | .py | # SPDX-License-Identifier: Apache-2.0
"""Zero-copy tensor codec for ZMQ multipart messages.
Frame 0: JSON metadata (tensor descriptors + scalar fields)
Frame 1-N: Raw tensor data buffers (one per tensor)
"""
import ctypes
import json
import logging
from dataclasses import dataclass
import torch
import zmq
logger = ... | 199 | 6,255 |
sglang | python/sglang/multimodal_gen/runtime/layers/moe.py | .py | # Adapted from LingBot-Video (https://github.com/Robbyant/lingbot-video).
#
# SPDX-License-Identifier: Apache-2.0
from __future__ import annotations
import torch
import torch.nn.functional as F
from torch import nn
class LingBotVideoMLP(nn.Module):
def __init__(self, hidden_size: int, intermediate_size: int) ->... | 183 | 6,907 |
sglang | python/sglang/multimodal_gen/runtime/layers/utils.py | .py | # Copied and adapted from: https://github.com/hao-ai-lab/FastVideo
# SPDX-License-Identifier: Apache-2.0
# Adapted from vllm: https://github.com/vllm-project/vllm/blob/v0.7.3/vllm/model_executor/layers/utils.py
"""Utility methods for model layers."""
import inspect
from typing import Any, Callable, List, Optional
im... | 268 | 9,778 |
sglang | python/sglang/multimodal_gen/runtime/layers/fused_scale_shift_gate.py | .py | # SPDX-License-Identifier: Apache-2.0
from typing import Optional, Tuple
import torch
import torch.nn.functional as F
from sglang.multimodal_gen.runtime.layers.custom_op import CustomOp
from sglang.multimodal_gen.runtime.platforms import current_platform
_is_cuda = current_platform.is_cuda()
if _is_cuda:
from s... | 160 | 5,433 |
sglang | python/sglang/multimodal_gen/runtime/layers/parallel_conv.py | .py | import contextvars
import math
from contextlib import contextmanager
import torch
import torch.distributed as dist
import torch.nn as nn
import torch.nn.functional as F
from sglang.multimodal_gen.runtime.distributed.parallel_state import (
get_decode_parallel_group_coordinator,
get_decode_parallel_rank,
g... | 802 | 25,168 |
sglang | python/sglang/multimodal_gen/runtime/layers/linear.py | .py | # Copied and adapted from: https://github.com/hao-ai-lab/FastVideo
# SPDX-License-Identifier: Apache-2.0
# Adapted from vllm: https://github.com/vllm-project/vllm/blob/v0.7.3/vllm/model_executor/layers/linear.py
from abc import abstractmethod
import torch
import torch.distributed as dist
import torch.nn.functional a... | 1,142 | 44,593 |
sglang | python/sglang/multimodal_gen/runtime/layers/layernorm.py | .py | # Copied and adapted from: https://github.com/hao-ai-lab/FastVideo
# SPDX-License-Identifier: Apache-2.0
# Adapted from vllm: https://github.com/vllm-project/vllm/blob/v0.7.3/vllm/model_executor/layers/layernorm.py
"""Custom normalization layers."""
import os
from typing import Optional, Tuple, Union
import torch
im... | 1,159 | 40,300 |
sglang | python/sglang/multimodal_gen/runtime/layers/activation.py | .py | # Copied and adapted from: https://github.com/hao-ai-lab/FastVideo
# SPDX-License-Identifier: Apache-2.0
# Adapted from vllm: https://github.com/vllm-project/vllm/blob/v0.7.3/vllm/model_executor/layers/activation.py
"""Custom activation functions."""
import math
from typing import Any
import torch
import torch.nn as... | 163 | 5,134 |
sglang | python/sglang/multimodal_gen/runtime/layers/visual_embedding.py | .py | # Copied and adapted from: https://github.com/hao-ai-lab/FastVideo
# SPDX-License-Identifier: Apache-2.0
import math
import torch
import torch.nn as nn
import torch.nn.functional as F
from diffusers.models.embeddings import (
CombinedTimestepGuidanceTextProjEmbeddings as _CombinedTimestepGuidanceTextProjEmbeddin... | 354 | 10,995 |
sglang | python/sglang/multimodal_gen/runtime/layers/custom_op.py | .py | # Copied and adapted from: https://github.com/hao-ai-lab/FastVideo
# SPDX-License-Identifier: Apache-2.0
# Adapted from vllm: https://github.com/vllm-project/vllm/blob/v0.7.3/vllm/model_executor/custom_op.py
from collections.abc import Callable
from typing import Any
import torch.nn as nn
from sglang.kernel_api_log... | 116 | 3,950 |
sglang | python/sglang/multimodal_gen/runtime/layers/mlp.py | .py | # Copied and adapted from: https://github.com/hao-ai-lab/FastVideo
# SPDX-License-Identifier: Apache-2.0
from typing import Optional
import torch
import torch.nn as nn
from diffusers.models.activations import (
GEGLU,
GELU,
ApproximateGELU,
LinearActivation,
SwiGLU,
)
from sglang.multimodal_gen.... | 122 | 3,831 |
sglang | python/sglang/multimodal_gen/runtime/layers/elementwise.py | .py | import torch
from sglang.kernels.ops.diffusion.triton.scale_shift import fuse_scale_shift_kernel
from sglang.multimodal_gen.runtime.layers.custom_op import CustomOp
class MulAdd(CustomOp):
"""
Fuse elementwise mul and add
Input: a, b, c, OptionalInt[k]
Output: a * (k + b) + c
"""
def __init_... | 54 | 1,692 |
sglang | python/sglang/multimodal_gen/runtime/layers/usp.py | .py | # Copied and adapted from: https://github.com/hao-ai-lab/FastVideo
import logging
from typing import TYPE_CHECKING
import torch
import torch.distributed as dist
import torch.distributed._functional_collectives as ft_c
from torch.distributed.tensor.experimental._attention import _cp_options
from sglang.kernels.ops.at... | 928 | 35,254 |
sglang | python/sglang/multimodal_gen/runtime/layers/vocab_parallel_embedding.py | .py | # Copied and adapted from: https://github.com/hao-ai-lab/FastVideo
# SPDX-License-Identifier: Apache-2.0
from collections.abc import Sequence
from dataclasses import dataclass
import torch
import torch.distributed as dist
import torch.nn.functional as F
from torch.nn.parameter import Parameter, UninitializedParamete... | 491 | 19,739 |
sglang | python/sglang/multimodal_gen/runtime/layers/kvcache/qvg_packed_cache.py | .py | # SPDX-License-Identifier: Apache-2.0
"""Quant-VideoGen KV cache with PRQ-packed storage for completed chunks.
Storage model (mirrors Quant-VideoGen's ChunkedKVCache, fitted to SGLang's
``update_and_get_attention_kv`` contract):
* The retained window is split into frame/chunk-aligned *segments* in global
token ... | 402 | 15,396 |
sglang | python/sglang/multimodal_gen/runtime/layers/kvcache/__init__.py | .py | # SPDX-License-Identifier: Apache-2.0
from sglang.multimodal_gen.runtime.layers.kvcache.causal_attention_cache import (
CausalAttentionKVView,
CausalSelfAttentionKVCache,
CrossAttentionKVCache,
)
__all__ = [
"CausalAttentionKVView",
"CausalSelfAttentionKVCache",
"CrossAttentionKVCache",
]
| 14 | 316 |
sglang | python/sglang/multimodal_gen/runtime/layers/kvcache/causal_attention_cache.py | .py | # SPDX-License-Identifier: Apache-2.0
from dataclasses import dataclass
import torch
@dataclass(slots=True)
class CausalAttentionKVView:
k: torch.Tensor
v: torch.Tensor
local_start_index: int
local_end_index: int
visible_local_end: int
visible_global_end: int
@dataclass(slots=True)
class C... | 534 | 21,030 |
sglang | python/sglang/multimodal_gen/runtime/layers/quantization/modelslim_mxfp8_scheme.py | .py | """ModelSlim MXFP8 scheme for pre-quantized weight inference on Ascend NPU.
Loads weights pre-quantized by msmodelslim (float8_e4m3fn weights,
uint8 scales) and runs MXFP8 matmul at inference.
"""
from typing import List, Optional
import torch
from sglang.multimodal_gen.runtime.platforms import current_platform
_i... | 125 | 4,482 |
sglang | python/sglang/multimodal_gen/runtime/layers/quantization/bitsandbytes.py | .py | # SPDX-License-Identifier: Apache-2.0
from __future__ import annotations
from typing import Any, Optional
import torch
import torch.nn as nn
from packaging import version
from sglang.multimodal_gen.runtime.layers.linear import (
LinearBase,
LinearMethodBase,
UnquantizedLinearMethod,
)
from sglang.multim... | 438 | 15,246 |
sglang | python/sglang/multimodal_gen/runtime/layers/quantization/modelopt_fp8.py | .py | """ModelOpt FP8 quantization support for diffusion models.
Handles checkpoints produced by NVIDIA Model Optimizer (ModelOpt) with
``quant_algo: "FP8"`` and ``quant_method: "modelopt"``.
Per quantized linear layer the checkpoint contains:
.weight float8_e4m3fn [out, in] FP8 quantized weight
.weight_... | 211 | 7,478 |
sglang | python/sglang/multimodal_gen/runtime/layers/quantization/modelslim.py | .py | from __future__ import annotations
import logging
from types import MappingProxyType
from typing import TYPE_CHECKING, Any, Dict, List, Mapping, Optional, cast
import torch
from sglang.multimodal_gen.runtime.layers.linear import (
LinearMethodBase,
UnquantizedLinearMethod,
)
from sglang.multimodal_gen.runtim... | 254 | 9,036 |
sglang | python/sglang/multimodal_gen/runtime/layers/quantization/fp8.py | .py | from __future__ import annotations
import logging
from typing import TYPE_CHECKING, Any, Dict, List, Optional, Union
import torch
from torch.nn import Module
from torch.nn.parameter import Parameter
from sglang.kernels.ops.quantization.fp8_kernel import (
is_fp8_fnuz,
per_token_group_quant_fp8,
)
from sglang... | 516 | 20,841 |
sglang | python/sglang/multimodal_gen/runtime/layers/quantization/mxfp4_npu.py | .py | """Online MXFP4 quantization for Diffusion models on Ascend NPU.
Provides ``NPUMXFP4Config`` (registered as ``"mxfp4_npu"``) and
``NPUMXFP4DiffusionLinearMethod`` which quantises FP16/BF16 weights to MXFP4
at load time using dual-level MX quantization and uses
``npu_dynamic_dual_level_mx_quant`` + ``npu_dual_level_qua... | 202 | 7,563 |
sglang | python/sglang/multimodal_gen/runtime/layers/quantization/__init__.py | .py | # Copied and adapted from: https://github.com/hao-ai-lab/FastVideo
from typing import Literal, get_args
from sglang.multimodal_gen.runtime.layers.quantization.bitsandbytes import (
BitsAndBytesConfig,
)
from sglang.multimodal_gen.runtime.layers.quantization.configs.base_config import (
QuantizationConfig,
)
f... | 99 | 3,160 |
sglang | python/sglang/multimodal_gen/runtime/layers/quantization/weight_only_fp8.py | .py | # SPDX-License-Identifier: Apache-2.0
import logging
import torch
import torch.nn as nn
import torch.nn.functional as F
import sglang.multimodal_gen.envs as envs
from sglang.multimodal_gen.runtime.distributed import (
divide,
get_tp_group,
split_tensor_along_last_dim,
tensor_model_parallel_all_gather... | 514 | 18,376 |
sglang | python/sglang/multimodal_gen/runtime/layers/quantization/modelopt_quant.py | .py | # Adapted from https://github.com/sgl-project/sglang/blob/main/python/sglang/srt/layers/quantization/modelopt_quant.py
from __future__ import annotations
import logging
import re
from functools import lru_cache
from typing import Any, Dict, List, Optional
import torch
from sglang.multimodal_gen.runtime.layers.linear... | 672 | 24,738 |
sglang | python/sglang/multimodal_gen/runtime/layers/quantization/modelslim_mxfp4_scheme.py | .py | """ModelSlim MXFP4 scheme for pre-quantized weight inference on Ascend NPU.
Loads weights pre-quantized by msmodelslim and runs MXFP4 dual-level
matmul at inference via npu_dual_level_quant_matmul.
Checkpoint tensor formats (verified from msmodelslim export):
weight: [out, in] float8_e4m3fn (FP... | 198 | 7,992 |
sglang | python/sglang/multimodal_gen/runtime/layers/quantization/mxfp4.py | .py | import logging
from typing import Dict, List, Optional
import torch
from torch.nn.parameter import Parameter
from sglang.multimodal_gen.runtime.layers.linear import (
LinearMethodBase,
UnquantizedLinearMethod,
)
from sglang.multimodal_gen.runtime.layers.quantization.configs.base_config import (
Quantizati... | 239 | 8,243 |
sglang | python/sglang/multimodal_gen/runtime/layers/quantization/mxfp8_npu.py | .py | """Online MXFP8 quantization for Diffusion models on Ascend NPU.
Provides ``MXFP8Config`` (registered as ``"mxfp8"``) and
``NPUMXFP8DiffusionLinearMethod`` which quantise FP16/BF16 weights to MXFP8
at load time and use ``npu_dynamic_mx_quant`` + ``npu_quant_matmul`` for
inference, mirroring the LLM-side ``NPUMXFP8Line... | 177 | 6,136 |
sglang | python/sglang/multimodal_gen/runtime/layers/quantization/nunchaku_linear.py | .py | # SPDX-License-Identifier: Apache-2.0
from typing import List, Optional
import torch
import torch.nn as nn
from torch.nn.parameter import Parameter
from sglang.multimodal_gen.runtime.layers.linear import LinearMethodBase
from sglang.multimodal_gen.runtime.utils.logging_utils import init_logger
from sglang.multimodal_... | 292 | 10,515 |
sglang | python/sglang/multimodal_gen/runtime/layers/quantization/configs/base_config.py | .py | # Copied and adapted from: https://github.com/hao-ai-lab/FastVideo
# SPDX-License-Identifier: Apache-2.0
# Adapted from vllm: https://github.com/vllm-project/vllm/blob/v0.7.3/vllm/model_executor/layers/quantization/base_config.py
import inspect
from abc import ABC, abstractmethod
from typing import TYPE_CHECKING, Any... | 156 | 5,323 |
sglang | python/sglang/multimodal_gen/runtime/layers/quantization/configs/nunchaku_config.py | .py | # SPDX-License-Identifier: Apache-2.0
import json
import os
from dataclasses import dataclass
from functools import lru_cache
from typing import Any, Optional
import torch
from safetensors.torch import load_file as safetensors_load_file
from torch import nn
from sglang.multimodal_gen.runtime.layers.linear import Line... | 284 | 9,321 |
sglang | python/sglang/multimodal_gen/runtime/layers/rotary_embedding/utils.py | .py | """Primitive RoPE ops: rotate helpers and apply_rotary_emb utilities."""
from typing import Optional, Tuple
import torch
from sglang.kernel_api_logging import debug_kernel_api
from sglang.kernels.ops.diffusion.triton.rotary import apply_rotary_embedding
from sglang.multimodal_gen.runtime.platforms import current_pla... | 197 | 7,321 |
sglang | python/sglang/multimodal_gen/runtime/layers/rotary_embedding/__init__.py | .py | # Copied and adapted from: https://github.com/hao-ai-lab/FastVideo
# SPDX-License-Identifier: Apache-2.0
# Adapted from vllm: https://github.com/vllm-project/vllm/blob/v0.7.3/vllm/model_executor/layers/rotary_embedding.py
# Adapted from
# https://github.com/huggingface/transformers/blob/v4.33.2/src/transformers/model... | 55 | 1,949 |
sglang | python/sglang/multimodal_gen/runtime/layers/rotary_embedding/factory.py | .py | """get_rope / get_rotary_pos_embed factory functions and module-level caches."""
from collections import OrderedDict
from typing import Any
import torch
from .base import LinearScalingRotaryEmbedding, RotaryEmbedding
from .mrope import NDRotaryEmbedding, _to_tuple
_ROPE_DICT: dict[tuple, RotaryEmbedding] = {}
_ND_R... | 172 | 5,645 |
sglang | python/sglang/multimodal_gen/runtime/layers/rotary_embedding/mrope.py | .py | """MRotaryEmbedding, YaRNScalingMRotaryEmbedding, NDRotaryEmbedding, OneDRotaryEmbedding."""
import functools
import torch
from sglang.multimodal_gen.runtime.distributed.parallel_state import get_sp_group
def _to_tuple(x: int | tuple[int, ...], dim: int = 2) -> tuple[int, ...]:
if isinstance(x, int):
r... | 503 | 19,659 |
sglang | python/sglang/multimodal_gen/runtime/layers/rotary_embedding/base.py | .py | """RotaryEmbedding base class and LinearScalingRotaryEmbedding variant."""
import torch
from sglang.multimodal_gen.runtime.layers.custom_op import CustomOp
from .utils import _apply_rotary_emb
@CustomOp.register("rotary_embedding")
class RotaryEmbedding(CustomOp):
"""Original rotary positional embedding."""
... | 134 | 4,723 |
sglang | python/sglang/multimodal_gen/runtime/layers/lora/linear.py | .py | # Copied and adapted from: https://github.com/hao-ai-lab/FastVideo
# SPDX-License-Identifier: Apache-2.0
# Code adapted from SGLang https://github.com/sgl-project/sglang/blob/main/python/sglang/srt/lora/layers.py
import os
import torch
from torch import nn
from torch.distributed._composable.fsdp import (
CPUOfflo... | 771 | 28,441 |
sglang | python/sglang/multimodal_gen/runtime/layers/attention/layer.py | .py | # Copied and adapted from: https://github.com/hao-ai-lab/FastVideo
# SPDX-License-Identifier: Apache-2.0
import functools
import os
from collections.abc import Sequence
from contextlib import nullcontext
from typing import Type
import torch
import torch.nn as nn
from torch.nn.attention import SDPBackend, sdpa_kernel
... | 1,773 | 71,832 |
sglang | python/sglang/multimodal_gen/runtime/layers/attention/__init__.py | .py | # Copied and adapted from: https://github.com/hao-ai-lab/FastVideo
# SPDX-License-Identifier: Apache-2.0
from sglang.multimodal_gen.runtime.layers.attention.backends.attention_backend import (
AttentionBackend,
AttentionMetadata,
AttentionMetadataBuilder,
)
from sglang.multimodal_gen.runtime.layers.attent... | 39 | 1,142 |
sglang | python/sglang/multimodal_gen/runtime/layers/attention/turbo_layer.py | .py | # copy and modify from https://github.com/thu-ml/TurboDiffusion/blob/main/turbodiffusion/rcm/utils/a2a_cp.py and https://github.com/thu-ml/TurboDiffusion/blob/main/turbodiffusion/SLA/core.py
from typing import Any, Callable, List, Tuple, Type, Union
import torch
import torch.distributed as dist
from einops import rea... | 326 | 11,869 |
sglang | python/sglang/multimodal_gen/runtime/layers/attention/selector.py | .py | # Copied and adapted from: https://github.com/hao-ai-lab/FastVideo
# SPDX-License-Identifier: Apache-2.0
# Adapted from vllm: https://github.com/vllm-project/vllm/blob/v0.7.3/vllm/attention/selector.py
import os
from collections.abc import Generator
from contextlib import contextmanager
from contextvars import Contex... | 381 | 13,391 |
sglang | python/sglang/multimodal_gen/runtime/layers/attention/STA_configuration.py | .py | # Copied and adapted from: https://github.com/hao-ai-lab/FastVideo
# SPDX-License-Identifier: Apache-2.0
import json
import os
from collections import defaultdict
from typing import Any
import numpy as np
from sglang.multimodal_gen.utils import dict_to_3d_list
def configure_sta(
mode: str = "STA_searching",
... | 415 | 16,484 |
sglang | python/sglang/multimodal_gen/runtime/layers/attention/backends/laser_attn.py | .py | import torch
from sglang.multimodal_gen.runtime.layers.attention.backends.attention_backend import (
AttentionBackend,
AttentionImpl,
AttentionMetadata,
)
from sglang.multimodal_gen.runtime.layers.attention.backends.sdpa import SDPABackend
from sglang.multimodal_gen.runtime.platforms import AttentionBacken... | 192 | 5,773 |
sglang | python/sglang/multimodal_gen/runtime/layers/attention/backends/ascend_fa.py | .py | from dataclasses import dataclass
from typing import Any
import torch
from sglang.multimodal_gen.runtime.layers.attention.backends.attention_backend import (
AttentionBackend,
AttentionImpl,
AttentionMetadata,
AttentionMetadataBuilder,
)
from sglang.multimodal_gen.runtime.platforms import AttentionBac... | 111 | 2,969 |
sglang | python/sglang/multimodal_gen/runtime/layers/attention/backends/rain_fusion_attn.py | .py | import math
from dataclasses import dataclass
from typing import Any, List, Optional
import attentions # noqa: F401
import torch
from einops import rearrange
from sglang.multimodal_gen.runtime.layers.attention.backends.attention_backend import (
AttentionBackend,
AttentionImpl,
AttentionMetadata,
Att... | 415 | 13,622 |
sglang | python/sglang/multimodal_gen/runtime/layers/attention/backends/xpu_backend.py | .py | # Copied and adapted from: https://github.com/hao-ai-lab/FastVideo
# SPDX-License-Identifier: Apache-2.0
from functools import lru_cache
import torch
from sglang.multimodal_gen.runtime.platforms import AttentionBackendEnum
try:
from sgl_kernel.flash_attn import flash_attn_varlen_func
flash_attn_func = flas... | 123 | 3,473 |
sglang | python/sglang/multimodal_gen/runtime/layers/attention/backends/sage_attn3.py | .py | # Copied and adapted from: https://github.com/hao-ai-lab/FastVideo
# SPDX-License-Identifier: Apache-2.0
import torch
import torch.nn.functional as F
from sageattn3 import sageattn3_blackwell
from sglang.multimodal_gen.runtime.layers.attention.backends.attention_backend import (
AttentionBackend,
AttentionIm... | 93 | 2,974 |
sglang | python/sglang/multimodal_gen/runtime/layers/attention/backends/attention_backend.py | .py | # Copied and adapted from: https://github.com/hao-ai-lab/FastVideo
# SPDX-License-Identifier: Apache-2.0
# Adapted from vllm: https://github.com/vllm-project/vllm/blob/v0.7.3/vllm/attention/backends/abstract.py
from abc import ABC, abstractmethod
from dataclasses import dataclass, fields
from typing import TYPE_CHECK... | 219 | 6,246 |
sglang | python/sglang/multimodal_gen/runtime/layers/attention/backends/sdpa.py | .py | # Copied and adapted from: https://github.com/hao-ai-lab/FastVideo
# SPDX-License-Identifier: Apache-2.0
from contextlib import nullcontext
import torch
from torch.nn.attention import SDPBackend, sdpa_kernel
from sglang.multimodal_gen.runtime.layers.attention.backends.attention_backend import ( # FlashAttentionMet... | 257 | 8,295 |
sglang | python/sglang/multimodal_gen/runtime/layers/attention/backends/sparse_video_gen_2_attn.py | .py | """
Sparse Video Gen 2 (SAP) attention backend.
This is a baseline integration that wires the backend into the
attention framework.
Adapted from https://github.com/svg-project/Sparse-VideoGen/blob/main/svg/models/wan/attention.py
"""
from dataclasses import dataclass, field
from typing import Any
import torch
impor... | 563 | 18,056 |
sglang | python/sglang/multimodal_gen/runtime/layers/attention/backends/block_sparse_attn.py | .py | from dataclasses import dataclass
from typing import Any
import attentions # noqa: F401
import torch
from sglang.multimodal_gen.runtime.layers.attention.backends.attention_backend import (
AttentionBackend,
AttentionImpl,
AttentionMetadata,
AttentionMetadataBuilder,
)
from sglang.multimodal_gen.runti... | 280 | 8,745 |
sglang | python/sglang/multimodal_gen/runtime/layers/attention/backends/video_sparse_attn.py | .py | # Copied and adapted from: https://github.com/hao-ai-lab/FastVideo
# SPDX-License-Identifier: Apache-2.0
import functools
import math
from dataclasses import dataclass
import torch
try:
from vsa import video_sparse_attn
except ImportError:
video_sparse_attn = None
from typing import Any
from sglang.multimo... | 329 | 10,494 |
sglang | python/sglang/multimodal_gen/runtime/layers/attention/backends/flash_attn.py | .py | # Copied and adapted from: https://github.com/hao-ai-lab/FastVideo
# SPDX-License-Identifier: Apache-2.0
from dataclasses import dataclass
from typing import Any, List, Optional, Tuple
import torch
from sglang.kernels.ops.attention.flash_attention import flash_attn_varlen_func
from sglang.multimodal_gen.runtime.layer... | 475 | 15,569 |
sglang | python/sglang/multimodal_gen/runtime/layers/attention/backends/sliding_tile_attn.py | .py | # Copied and adapted from: https://github.com/hao-ai-lab/FastVideo
# SPDX-License-Identifier: Apache-2.0
import json
from dataclasses import dataclass
from typing import Any
import torch
from einops import rearrange
from sglang.multimodal_gen.runtime.distributed import get_sp_group
from sglang.multimodal_gen.runtime... | 317 | 10,801 |
sglang | python/sglang/multimodal_gen/runtime/layers/attention/backends/aiter_sage.py | .py | # SPDX-License-Identifier: Apache-2.0
import torch
from sglang.multimodal_gen.runtime.layers.attention.backends.attention_backend import (
AttentionBackend,
AttentionImpl,
AttentionMetadata,
AttentionMetadataBuilder,
)
from sglang.multimodal_gen.runtime.platforms import AttentionBackendEnum
class A... | 82 | 2,353 |
sglang | python/sglang/multimodal_gen/runtime/layers/attention/backends/sol_attn.py | .py | # SPDX-License-Identifier: Apache-2.0
from __future__ import annotations
import re
import torch
from sglang.kernels.ops.attention.flash_attention import flash_attn_varlen_func
from sglang.multimodal_gen.runtime.layers.attention.backends.attention_backend import (
AttentionBackend,
AttentionImpl,
Attenti... | 241 | 7,583 |
sglang | python/sglang/multimodal_gen/runtime/layers/attention/backends/vmoba.py | .py | # Copied and adapted from: https://github.com/hao-ai-lab/FastVideo
# SPDX-License-Identifier: Apache-2.0
import re
from dataclasses import dataclass
import torch
from einops import rearrange
from kernel.attn.vmoba_attn.vmoba import (
moba_attn_varlen,
process_moba_input,
process_moba_output,
)
from sgla... | 260 | 8,337 |
sglang | python/sglang/multimodal_gen/runtime/layers/attention/backends/sparse_linear_attn.py | .py | """
Copyright (c) 2025 by SLA team.
Licensed under the Apache License, Version 2.0 (the "License");
This implementation is adapted from: from https://github.com/thu-ml/TurboDiffusion/blob/main/turbodiffusion/SLA/core.py and https://github.com/thu-ml/SLA/blob/main/SageSLA/core.py
Citation (please cite if you use this ... | 579 | 19,065 |
sglang | python/sglang/multimodal_gen/runtime/layers/attention/backends/flash_attn_2.py | .py | # Copied and adapted from: https://github.com/hao-ai-lab/FastVideo
# SPDX-License-Identifier: Apache-2.0
import torch
from sglang.multimodal_gen.runtime.layers.attention.backends.attention_backend import (
AttentionBackend,
AttentionImpl,
AttentionMetadata,
AttentionMetadataBuilder,
)
from sglang.mul... | 80 | 2,121 |
sglang | python/sglang/multimodal_gen/runtime/layers/attention/backends/sage_attn.py | .py | # Copied and adapted from: https://github.com/hao-ai-lab/FastVideo
# SPDX-License-Identifier: Apache-2.0
import torch
from sageattention import sageattn
from sglang.multimodal_gen.runtime.layers.attention.backends.attention_backend import ( # FlashAttentionMetadata,
AttentionBackend,
AttentionImpl,
Att... | 160 | 4,581 |
sglang | python/sglang/multimodal_gen/runtime/layers/attention/backends/subblock_sparse_attn.py | .py | # SPDX-License-Identifier: Apache-2.0
"""SubBlock block-sparse attention backend.
Routes FlashInfer's 64-token block-sparse kernel with a K-side sub-block
log-sum-exp score (see ``backends/subblock_sparse/``). Everything is training-free:
the router runs before attention and produces the ``q2k_block_index`` the
kernel... | 408 | 16,262 |
sglang | python/sglang/multimodal_gen/runtime/layers/attention/backends/aiter.py | .py | # Copied and adapted from: https://github.com/hao-ai-lab/FastVideo
# SPDX-License-Identifier: Apache-2.0
import importlib
import logging
import os
import aiter
import torch
from sglang.multimodal_gen.runtime.layers.attention.backends.attention_backend import (
AttentionBackend,
AttentionImpl,
AttentionM... | 247 | 7,707 |
sglang | python/sglang/multimodal_gen/runtime/layers/attention/backends/subblock_sparse/kernels.py | .py | # SPDX-License-Identifier: Apache-2.0
"""Fused sub-block score kernel: bf16 GEMM + segmented log-sum-exp in one pass.
score[i, j] = log2 sum_{b < n_k} 2 ** ( qbar_i . kbar_{j,b} )
The PyTorch router materialises ``[B, H, Gq, Gk*n_k]`` fp32 (254 MB at S=96k, n_k=4),
reduces it, and throws it away -- the GEMM is 4%... | 301 | 10,512 |
sglang | python/sglang/multimodal_gen/runtime/layers/attention/backends/subblock_sparse/__init__.py | .py | # SPDX-License-Identifier: Apache-2.0
"""SubBlock -- training-free block-sparse attention routing for video DiTs.
Originally vendored from the standalone SubBlock repository; ``router.py`` and
``kernels.py`` have since diverged from it.
``router.py`` scores every (query block, key block) pair from sub-block-pooled
Q/... | 16 | 650 |
sglang | python/sglang/multimodal_gen/runtime/layers/attention/backends/subblock_sparse/router.py | .py | # SPDX-License-Identifier: Apache-2.0
"""Sub-block block-sparse routing for FlashInfer's ``bsa_attn_blk64_fwd``.
Training-free. Runs *before* attention, produces the ``q2k_block_index`` tensor the
64-token block-sparse kernel consumes.
Why sub-blocks
--------------
The usual proxy score for a 64x64 block is ``mean(Q_... | 257 | 11,811 |
sglang | python/sglang/multimodal_gen/runtime/entrypoints/utils.py | .py | # Copied and adapted from: https://github.com/hao-ai-lab/FastVideo
# SPDX-License-Identifier: Apache-2.0
"""
DiffGenerator module for sglang-diffusion.
This module provides a consolidated interface for generating videos using
diffusion models.
"""
import atexit
import json
import mmap
import os
import shutil
import ... | 1,288 | 41,862 |
sglang | python/sglang/multimodal_gen/runtime/entrypoints/diffusion_generator.py | .py | # Copied and adapted from: https://github.com/hao-ai-lab/FastVideo
# SPDX-License-Identifier: Apache-2.0
"""
DiffGenerator module for sglang-diffusion.
This module provides a consolidated interface for generating images/videos using
diffusion models.
"""
import dataclasses
import multiprocessing as mp
import os
impo... | 759 | 30,351 |
sglang | python/sglang/multimodal_gen/runtime/entrypoints/http_server.py | .py | # Copied and adapted from: https://github.com/hao-ai-lab/FastVideo
import asyncio
import base64
import os
import signal
import uuid
from contextlib import asynccontextmanager, suppress
from typing import TYPE_CHECKING
import httpx
import torch
from fastapi import APIRouter, FastAPI, Request, Response
from fastapi.mid... | 436 | 14,767 |
sglang | python/sglang/multimodal_gen/runtime/entrypoints/post_training/io_struct.py | .py | """Request/response data structures for post-training APIs."""
from __future__ import annotations
from dataclasses import dataclass
from typing import Any, Optional
from pydantic import BaseModel
@dataclass
class UpdateWeightFromDiskReqInput:
"""Request to update model weights from disk for diffusion models.""... | 116 | 3,206 |
sglang | python/sglang/multimodal_gen/runtime/entrypoints/post_training/utils.py | .py | """Tensor serialization for post-training / rollout HTTP responses."""
from __future__ import annotations
from typing import Any
import numpy as np
import torch
from safetensors.torch import load, save
def tensor_to_bytes(t: torch.Tensor) -> bytes:
return save({"t": t.detach().contiguous().cpu()})
def bytes_... | 45 | 1,252 |
sglang | python/sglang/multimodal_gen/runtime/entrypoints/post_training/rollout_api.py | .py | """Rollout HTTP API (``POST /rollout/generate``)."""
from __future__ import annotations
from typing import Any
import msgspec
import torch
from fastapi import APIRouter, HTTPException, Response
from sglang.multimodal_gen.configs.sample.sampling_params import generate_request_id
from sglang.multimodal_gen.runtime.en... | 351 | 13,024 |
sglang | python/sglang/multimodal_gen/runtime/entrypoints/post_training/weights_api.py | .py | """Weight update API for the diffusion engine."""
from fastapi import APIRouter, Request
from sglang.multimodal_gen.runtime.entrypoints.post_training.io_struct import (
GetWeightsChecksumReqInput,
ReleaseMemoryOccupationReqInput,
ResumeMemoryOccupationReqInput,
UpdateWeightFromDiskReqInput,
Update... | 204 | 6,438 |
sglang | python/sglang/multimodal_gen/runtime/entrypoints/openai/stores.py | .py | import asyncio
from typing import Any, Dict, Optional
class AsyncDictStore:
"""A small async-safe in-memory key-value store for dict items.
This encapsulates the usual pattern of a module-level dict guarded by
an asyncio.Lock and provides simple CRUD methods that are safe to call
concurrently from Fa... | 71 | 2,248 |
sglang | python/sglang/multimodal_gen/runtime/entrypoints/openai/video_api.py | .py | # Copied and adapted from: https://github.com/hao-ai-lab/FastVideo
import asyncio
import json
import os
import shutil
import tempfile
import time
from collections.abc import Coroutine
from contextlib import suppress
from typing import Any, Dict, Optional
from fastapi import (
APIRouter,
File,
Form,
HT... | 965 | 35,717 |
sglang | python/sglang/multimodal_gen/runtime/entrypoints/openai/utils.py | .py | # Copied and adapted from: https://github.com/hao-ai-lab/FastVideo
import asyncio
import inspect
import json
import os
import shutil
import tempfile
import time
from contextlib import contextmanager
from typing import Any, Generator, List, Optional, Union
import httpx
from fastapi import HTTPException, UploadFile
fro... | 472 | 16,768 |
sglang | python/sglang/multimodal_gen/runtime/entrypoints/openai/mesh_api.py | .py | import asyncio
import os
import time
from typing import Any, Dict, List, Optional
from fastapi import (
APIRouter,
File,
Form,
HTTPException,
Path,
Query,
Request,
UploadFile,
)
from fastapi.responses import FileResponse
from sglang.multimodal_gen.configs.sample.sampling_params import ... | 281 | 9,560 |
sglang | python/sglang/multimodal_gen/runtime/entrypoints/openai/storage.py | .py | import asyncio
import os
from typing import Optional
from sglang.multimodal_gen.runtime.utils.logging_utils import init_logger
logger = init_logger(__name__)
class CloudStorage:
def __init__(self):
self.enabled = os.getenv("SGLANG_CLOUD_STORAGE_TYPE", "").lower() == "s3"
if not self.enabled:
... | 110 | 3,635 |
sglang | python/sglang/multimodal_gen/runtime/entrypoints/openai/image_api.py | .py | # Copied and adapted from: https://github.com/hao-ai-lab/FastVideo
import asyncio
import base64
import contextlib
import json
import os
import time
from typing import Any, List, Optional
from fastapi import (
APIRouter,
File,
Form,
HTTPException,
Path,
Query,
Request,
UploadFile,
)
fro... | 598 | 21,620 |
sglang | python/sglang/multimodal_gen/runtime/entrypoints/openai/protocol.py | .py | import time
import uuid
from abc import ABC
from dataclasses import dataclass, field
from typing import Any, Dict, List, Literal, Optional, Union
from pydantic import BaseModel, ConfigDict, Field
# Image API protocol models
class ImageResponseData(BaseModel):
b64_json: Optional[str] = None
url: Optional[str]... | 246 | 8,182 |
sglang | python/sglang/multimodal_gen/runtime/entrypoints/openai/common_api.py | .py | import time
from typing import Any, List, Optional, Union
from fastapi import APIRouter, Body, HTTPException
from pydantic import BaseModel, Field
from sglang.multimodal_gen.registry import get_model_info
from sglang.multimodal_gen.runtime.entrypoints.utils import (
ListLorasReq,
MergeLoraWeightsReq,
SetL... | 239 | 8,746 |
sglang | python/sglang/multimodal_gen/runtime/entrypoints/openai/realtime/generate_session.py | .py | # SPDX-License-Identifier: Apache-2.0
from __future__ import annotations
from dataclasses import dataclass
from typing import TYPE_CHECKING, Any
from uuid import uuid4
from sglang.multimodal_gen.runtime.entrypoints.openai.protocol import (
RealtimeVideoGenerationsRequest,
)
from sglang.multimodal_gen.runtime.rea... | 85 | 2,648 |
sglang | python/sglang/multimodal_gen/runtime/entrypoints/openai/realtime/realtime_adapter.py | .py | # SPDX-License-Identifier: Apache-2.0
from __future__ import annotations
import hashlib
import os
import tempfile
from dataclasses import dataclass, field
from typing import TYPE_CHECKING, Any
from fastapi import WebSocket
from sglang.multimodal_gen.runtime.entrypoints.openai.protocol import (
RealtimeEvent,
... | 269 | 8,701 |
sglang | python/sglang/multimodal_gen/runtime/entrypoints/openai/realtime/registry.py | .py | # SPDX-License-Identifier: Apache-2.0
from __future__ import annotations
from typing import TYPE_CHECKING
from sglang.multimodal_gen.runtime.entrypoints.openai.realtime.realtime_adapter import (
BaseRealtimeModelAdapter,
)
if TYPE_CHECKING:
from sglang.multimodal_gen.runtime.server_args import ServerArgs
... | 70 | 2,132 |
sglang | python/sglang/multimodal_gen/runtime/entrypoints/openai/realtime/timer.py | .py | # SPDX-License-Identifier: Apache-2.0
import time
class RealtimeStageTimer:
__slots__ = ("_last", "_start")
def __init__(self):
now = time.perf_counter()
self._start = now
self._last = now
def mark_ms(self) -> float:
now = time.perf_counter()
elapsed_ms = (now - ... | 22 | 486 |
sglang | python/sglang/multimodal_gen/runtime/entrypoints/openai/realtime/realtime_output_adapter.py | .py | # SPDX-License-Identifier: Apache-2.0
from __future__ import annotations
import asyncio
import io
from dataclasses import dataclass
from typing import TYPE_CHECKING, TypedDict
import msgspec.msgpack
from fastapi import WebSocket
from PIL import Image
from sglang.multimodal_gen.runtime.entrypoints.openai.realtime.ti... | 611 | 19,606 |
sglang | python/sglang/multimodal_gen/runtime/entrypoints/openai/realtime/realtime_video_api.py | .py | # SPDX-License-Identifier: Apache-2.0
import asyncio
import shutil
import time
from typing import TYPE_CHECKING
import msgspec.msgpack
from fastapi import APIRouter, WebSocket, WebSocketDisconnect
from sglang.multimodal_gen.runtime.entrypoints.openai.protocol import (
RealtimeEvent,
RealtimeVideoGenerationsR... | 501 | 17,069 |
sglang | python/sglang/multimodal_gen/runtime/entrypoints/openai/realtime/adapters/lingbot_world_realtime_adapter.py | .py | # SPDX-License-Identifier: Apache-2.0
from __future__ import annotations
from typing import TYPE_CHECKING, Any
from sglang.multimodal_gen.runtime.entrypoints.openai.protocol import (
RealtimeEvent,
RealtimeVideoGenerationsRequest,
)
from sglang.multimodal_gen.runtime.entrypoints.openai.realtime.realtime_adap... | 353 | 12,453 |
sglang | python/sglang/multimodal_gen/runtime/entrypoints/openai/realtime/adapters/sana_wm_realtime_adapter.py | .py | # SPDX-License-Identifier: Apache-2.0
from __future__ import annotations
from typing import TYPE_CHECKING, Any
from sglang.multimodal_gen.runtime.entrypoints.openai.protocol import (
RealtimeEvent,
RealtimeVideoGenerationsRequest,
)
from sglang.multimodal_gen.runtime.entrypoints.openai.realtime.realtime_adap... | 265 | 9,942 |
sglang | python/sglang/multimodal_gen/runtime/entrypoints/action/openpi.py | .py | # SPDX-License-Identifier: Apache-2.0
from __future__ import annotations
from typing import Any
from fastapi import APIRouter, WebSocket
from sglang.multimodal_gen.runtime.entrypoints.action.ws_utils import (
run_action_msgpack_ws,
)
from sglang.multimodal_gen.runtime.server_args import ServerArgs
router = API... | 30 | 768 |
sglang | python/sglang/multimodal_gen/runtime/entrypoints/action/protocol.py | .py | # SPDX-License-Identifier: Apache-2.0
from __future__ import annotations
import base64
import dataclasses
import io
import time
import uuid
from functools import lru_cache
from typing import Any
import numpy as np
from PIL import Image
from sglang.multimodal_gen.configs.pipeline_configs.cosmos3 import Cosmos3Config... | 663 | 24,721 |
sglang | python/sglang/multimodal_gen/runtime/entrypoints/action/api.py | .py | # SPDX-License-Identifier: Apache-2.0
from __future__ import annotations
import json
import os
from contextlib import nullcontext
from typing import Any
from fastapi import APIRouter, HTTPException, Request, Response, WebSocket
from sglang.multimodal_gen.configs.sample.sampling_params import generate_request_id
fro... | 254 | 8,100 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.