Upload 2 files
Browse files- handler.py +79 -20
- requirements.txt +2 -3
handler.py
CHANGED
@@ -1,42 +1,101 @@
|
|
|
|
|
|
1 |
import os
|
2 |
from typing import Any, Dict
|
3 |
|
4 |
from diffusers import FluxPipeline, FluxTransformer2DModel, AutoencoderKL, TorchAoConfig
|
5 |
from PIL import Image
|
6 |
import torch
|
|
|
|
|
7 |
|
8 |
IS_COMPILE = False
|
|
|
|
|
9 |
|
10 |
if IS_COMPILE:
|
11 |
import torch._dynamo
|
12 |
torch._dynamo.config.suppress_errors = True
|
13 |
|
14 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
15 |
|
16 |
-
def
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
17 |
pipe.transformer.to(memory_format=torch.channels_last)
|
18 |
-
pipe.transformer = torch.compile(pipe.transformer, mode="reduce-overhead", fullgraph=False, dynamic=False
|
19 |
-
|
20 |
-
|
|
|
21 |
return pipe
|
22 |
|
23 |
class EndpointHandler:
|
24 |
def __init__(self, path=""):
|
25 |
-
repo_id = "
|
26 |
-
#
|
27 |
-
dtype = torch.
|
28 |
-
|
29 |
-
|
30 |
-
|
31 |
-
self.pipeline = FluxPipeline.from_pretrained(repo_id, vae=vae, torch_dtype=dtype, quantization_config=quantization_config)
|
32 |
-
self.pipeline.transformer.fuse_qkv_projections()
|
33 |
-
self.pipeline.vae.fuse_qkv_projections()
|
34 |
-
if IS_COMPILE: self.pipeline = compile_pipeline(self.pipeline)
|
35 |
-
self.pipeline.to("cuda")
|
36 |
-
|
37 |
-
@torch.inference_mode()
|
38 |
def __call__(self, data: Dict[str, Any]) -> Image.Image:
|
39 |
-
|
40 |
|
41 |
if "inputs" in data and isinstance(data["inputs"], str):
|
42 |
prompt = data.pop("inputs")
|
@@ -50,7 +109,7 @@ class EndpointHandler:
|
|
50 |
|
51 |
parameters = data.pop("parameters", {})
|
52 |
|
53 |
-
num_inference_steps = parameters.get("num_inference_steps", 28)
|
54 |
width = parameters.get("width", 1024)
|
55 |
height = parameters.get("height", 1024)
|
56 |
guidance_scale = parameters.get("guidance_scale", 3.5)
|
|
|
1 |
+
# https://github.com/sayakpaul/diffusers-torchao
|
2 |
+
|
3 |
import os
|
4 |
from typing import Any, Dict
|
5 |
|
6 |
from diffusers import FluxPipeline, FluxTransformer2DModel, AutoencoderKL, TorchAoConfig
|
7 |
from PIL import Image
|
8 |
import torch
|
9 |
+
from torchao.quantization import quantize_, autoquant, int8_dynamic_activation_int8_weight, int8_dynamic_activation_int4_weight
|
10 |
+
from huggingface_hub import hf_hub_download
|
11 |
|
12 |
IS_COMPILE = False
|
13 |
+
IS_TURBO = False
|
14 |
+
IS_4BIT = True
|
15 |
|
16 |
if IS_COMPILE:
|
17 |
import torch._dynamo
|
18 |
torch._dynamo.config.suppress_errors = True
|
19 |
|
20 |
+
from huggingface_inference_toolkit.logging import logger
|
21 |
+
|
22 |
+
def load_pipeline_stable(repo_id: str, dtype: torch.dtype) -> Any:
|
23 |
+
quantization_config = TorchAoConfig("int4dq" if IS_4BIT else "int8dq")
|
24 |
+
vae = AutoencoderKL.from_pretrained(repo_id, subfolder="vae", torch_dtype=dtype)
|
25 |
+
pipe = FluxPipeline.from_pretrained(repo_id, vae=vae, torch_dtype=dtype, quantization_config=quantization_config)
|
26 |
+
pipe.transformer.fuse_qkv_projections()
|
27 |
+
pipe.vae.fuse_qkv_projections()
|
28 |
+
pipe.to("cuda")
|
29 |
+
return pipe
|
30 |
+
|
31 |
+
def load_pipeline_compile(repo_id: str, dtype: torch.dtype) -> Any:
|
32 |
+
quantization_config = TorchAoConfig("int4dq" if IS_4BIT else "int8dq")
|
33 |
+
vae = AutoencoderKL.from_pretrained(repo_id, subfolder="vae", torch_dtype=dtype)
|
34 |
+
pipe = FluxPipeline.from_pretrained(repo_id, vae=vae, torch_dtype=dtype, quantization_config=quantization_config)
|
35 |
+
pipe.transformer.fuse_qkv_projections()
|
36 |
+
pipe.vae.fuse_qkv_projections()
|
37 |
+
pipe.transformer.to(memory_format=torch.channels_last)
|
38 |
+
pipe.transformer = torch.compile(pipe.transformer, mode="reduce-overhead", fullgraph=False, dynamic=False)
|
39 |
+
pipe.vae.to(memory_format=torch.channels_last)
|
40 |
+
pipe.vae = torch.compile(pipe.vae, mode="reduce-overhead", fullgraph=False, dynamic=False)
|
41 |
+
pipe.to("cuda")
|
42 |
+
return pipe
|
43 |
+
|
44 |
+
def load_pipeline_autoquant(repo_id: str, dtype: torch.dtype) -> Any:
|
45 |
+
pipe = FluxPipeline.from_pretrained(repo_id, torch_dtype=dtype).to("cuda")
|
46 |
+
pipe.transformer.fuse_qkv_projections()
|
47 |
+
pipe.vae.fuse_qkv_projections()
|
48 |
+
pipe.transformer.to(memory_format=torch.channels_last)
|
49 |
+
pipe.transformer = torch.compile(pipe.transformer, mode="max-autotune", fullgraph=True)
|
50 |
+
pipe.vae.to(memory_format=torch.channels_last)
|
51 |
+
pipe.vae = torch.compile(pipe.vae, mode="max-autotune", fullgraph=True)
|
52 |
+
pipe.transformer = autoquant(pipe.transformer, error_on_unseen=False)
|
53 |
+
pipe.vae = autoquant(pipe.vae, error_on_unseen=False)
|
54 |
+
pipe.to("cuda")
|
55 |
+
return pipe
|
56 |
+
|
57 |
+
def load_pipeline_turbo(repo_id: str, dtype: torch.dtype) -> Any:
|
58 |
+
pipe = FluxPipeline.from_pretrained(repo_id, torch_dtype=dtype).to("cuda")
|
59 |
+
pipe.load_lora_weights(hf_hub_download("ByteDance/Hyper-SD", "Hyper-FLUX.1-dev-8steps-lora.safetensors"), adapter_name="hyper-sd")
|
60 |
+
pipe.set_adapters(["hyper-sd"], adapter_weights=[0.125])
|
61 |
+
pipe.fuse_lora()
|
62 |
+
pipe.transformer.fuse_qkv_projections()
|
63 |
+
pipe.vae.fuse_qkv_projections()
|
64 |
+
weight = int8_dynamic_activation_int4_weight() if IS_4BIT else int8_dynamic_activation_int8_weight()
|
65 |
+
quantize_(pipe.transformer, weight, device="cuda")
|
66 |
+
quantize_(pipe.vae, weight, device="cuda")
|
67 |
+
quantize_(pipe.text_encoder_2, weight, device="cuda")
|
68 |
+
pipe.to("cuda")
|
69 |
+
return pipe
|
70 |
|
71 |
+
def load_pipeline_turbo_compile(repo_id: str, dtype: torch.dtype) -> Any:
|
72 |
+
pipe = FluxPipeline.from_pretrained(repo_id, torch_dtype=dtype).to("cuda")
|
73 |
+
pipe.load_lora_weights(hf_hub_download("ByteDance/Hyper-SD", "Hyper-FLUX.1-dev-8steps-lora.safetensors"), adapter_name="hyper-sd")
|
74 |
+
pipe.set_adapters(["hyper-sd"], adapter_weights=[0.125])
|
75 |
+
pipe.fuse_lora()
|
76 |
+
pipe.transformer.fuse_qkv_projections()
|
77 |
+
pipe.vae.fuse_qkv_projections()
|
78 |
+
weight = int8_dynamic_activation_int4_weight() if IS_4BIT else int8_dynamic_activation_int8_weight()
|
79 |
+
quantize_(pipe.transformer, weight, device="cuda")
|
80 |
+
quantize_(pipe.vae, weight, device="cuda")
|
81 |
+
quantize_(pipe.text_encoder_2, weight, device="cuda")
|
82 |
pipe.transformer.to(memory_format=torch.channels_last)
|
83 |
+
pipe.transformer = torch.compile(pipe.transformer, mode="reduce-overhead", fullgraph=False, dynamic=False)
|
84 |
+
pipe.vae.to(memory_format=torch.channels_last)
|
85 |
+
pipe.vae = torch.compile(pipe.vae, mode="reduce-overhead", fullgraph=False, dynamic=False)
|
86 |
+
pipe.to("cuda")
|
87 |
return pipe
|
88 |
|
89 |
class EndpointHandler:
|
90 |
def __init__(self, path=""):
|
91 |
+
repo_id = "NoMoreCopyrightOrg/flux-dev-8step" if IS_TURBO else "NoMoreCopyrightOrg/flux-dev"
|
92 |
+
#dtype = torch.bfloat16
|
93 |
+
dtype = torch.float16 # for older nVidia GPUs
|
94 |
+
if IS_COMPILE: load_pipeline_compile(repo_id, dtype)
|
95 |
+
else: self.pipeline = load_pipeline_stable(repo_id, dtype)
|
96 |
+
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
97 |
def __call__(self, data: Dict[str, Any]) -> Image.Image:
|
98 |
+
logger.info(f"Received incoming request with {data=}")
|
99 |
|
100 |
if "inputs" in data and isinstance(data["inputs"], str):
|
101 |
prompt = data.pop("inputs")
|
|
|
109 |
|
110 |
parameters = data.pop("parameters", {})
|
111 |
|
112 |
+
num_inference_steps = parameters.get("num_inference_steps", 8 if IS_TURBO else 28)
|
113 |
width = parameters.get("width", 1024)
|
114 |
height = parameters.get("height", 1024)
|
115 |
guidance_scale = parameters.get("guidance_scale", 3.5)
|
requirements.txt
CHANGED
@@ -1,15 +1,14 @@
|
|
1 |
huggingface_hub
|
2 |
torch==2.4.0
|
3 |
torchvision
|
|
|
4 |
torchao==0.9.0
|
5 |
-
diffusers
|
6 |
peft
|
7 |
-
accelerate
|
8 |
transformers
|
9 |
numpy
|
10 |
scipy
|
11 |
Pillow
|
12 |
sentencepiece
|
13 |
protobuf
|
14 |
-
pytorch-lightning
|
15 |
triton
|
|
|
1 |
huggingface_hub
|
2 |
torch==2.4.0
|
3 |
torchvision
|
4 |
+
torchaudio
|
5 |
torchao==0.9.0
|
6 |
+
diffusers==0.32.2
|
7 |
peft
|
|
|
8 |
transformers
|
9 |
numpy
|
10 |
scipy
|
11 |
Pillow
|
12 |
sentencepiece
|
13 |
protobuf
|
|
|
14 |
triton
|