diffusers 0.33.1__py3-none-any.whl → 0.35.0__py3-none-any.whl

diffusers/schedulers/scheduling_dpmsolver_multistep.py

@@ -1,4 +1,4 @@
-# Copyright 2024 TSAIL Team and The HuggingFace Team. All rights reserved.
+# Copyright 2025 TSAIL Team and The HuggingFace Team. All rights reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -78,7 +78,7 @@ def betas_for_alpha_bar(
 # Copied from diffusers.schedulers.scheduling_ddim.rescale_zero_terminal_snr
 def rescale_zero_terminal_snr(betas):
     """
-    Rescales betas to have zero terminal SNR Based on https://arxiv.org/pdf/2305.08891.pdf (Algorithm 1)
+    Rescales betas to have zero terminal SNR Based on https://huggingface.co/papers/2305.08891 (Algorithm 1)
 
 
     Args:
@@ -230,6 +230,8 @@ class DPMSolverMultistepScheduler(SchedulerMixin, ConfigMixin):
         timestep_spacing: str = "linspace",
         steps_offset: int = 0,
         rescale_betas_zero_snr: bool = False,
+        use_dynamic_shifting: bool = False,
+        time_shift_type: str = "exponential",
     ):
         if self.config.use_beta_sigmas and not is_scipy_available():
             raise ImportError("Make sure to install scipy if you want to use beta sigmas.")
@@ -330,6 +332,7 @@ class DPMSolverMultistepScheduler(SchedulerMixin, ConfigMixin):
         self,
         num_inference_steps: int = None,
         device: Union[str, torch.device] = None,
+        mu: Optional[float] = None,
         timesteps: Optional[List[int]] = None,
     ):
         """
@@ -345,6 +348,9 @@ class DPMSolverMultistepScheduler(SchedulerMixin, ConfigMixin):
                 based on the `timestep_spacing` attribute. If `timesteps` is passed, `num_inference_steps` and `sigmas`
                 must be `None`, and `timestep_spacing` attribute will be ignored.
         """
+        if mu is not None:
+            assert self.config.use_dynamic_shifting and self.config.time_shift_type == "exponential"
+            self.config.flow_shift = np.exp(mu)
         if num_inference_steps is None and timesteps is None:
             raise ValueError("Must pass exactly one of `num_inference_steps` or `timesteps`.")
         if num_inference_steps is not None and timesteps is not None:
@@ -366,7 +372,7 @@ class DPMSolverMultistepScheduler(SchedulerMixin, ConfigMixin):
             clipped_idx = torch.searchsorted(torch.flip(self.lambda_t, [0]), self.config.lambda_min_clipped)
             last_timestep = ((self.config.num_train_timesteps - clipped_idx).numpy()).item()
 
-            # "linspace", "leading", "trailing" corresponds to annotation of Table 2. of https://arxiv.org/abs/2305.08891
+            # "linspace", "leading", "trailing" corresponds to annotation of Table 2. of https://huggingface.co/papers/2305.08891
             if self.config.timestep_spacing == "linspace":
                 timesteps = (
                     np.linspace(0, last_timestep - 1, num_inference_steps + 1)
@@ -460,7 +466,7 @@ class DPMSolverMultistepScheduler(SchedulerMixin, ConfigMixin):
         pixels from saturation at each step. We find that dynamic thresholding results in significantly better
         photorealism as well as better image-text alignment, especially when using very large guidance weights."
 
-        https://arxiv.org/abs/2205.11487
+        https://huggingface.co/papers/2205.11487
         """
         dtype = sample.dtype
         batch_size, channels, *remaining_dims = sample.shape
@@ -646,7 +652,7 @@ class DPMSolverMultistepScheduler(SchedulerMixin, ConfigMixin):
             if len(args) > 1:
                 sample = args[1]
             else:
-                raise ValueError("missing `sample` as a required keyward argument")
+                raise ValueError("missing `sample` as a required keyword argument")
         if timestep is not None:
             deprecate(
                 "timesteps",
@@ -741,7 +747,7 @@ class DPMSolverMultistepScheduler(SchedulerMixin, ConfigMixin):
             if len(args) > 2:
                 sample = args[2]
             else:
-                raise ValueError(" missing `sample` as a required keyward argument")
+                raise ValueError("missing `sample` as a required keyword argument")
         if timestep is not None:
             deprecate(
                 "timesteps",
@@ -810,7 +816,7 @@ class DPMSolverMultistepScheduler(SchedulerMixin, ConfigMixin):
             if len(args) > 2:
                 sample = args[2]
             else:
-                raise ValueError(" missing `sample` as a required keyward argument")
+                raise ValueError("missing `sample` as a required keyword argument")
         if timestep_list is not None:
             deprecate(
                 "timestep_list",
@@ -845,7 +851,7 @@ class DPMSolverMultistepScheduler(SchedulerMixin, ConfigMixin):
         r0 = h_0 / h
         D0, D1 = m0, (1.0 / r0) * (m0 - m1)
         if self.config.algorithm_type == "dpmsolver++":
-            # See https://arxiv.org/abs/2211.01095 for detailed derivations
+            # See https://huggingface.co/papers/2211.01095 for detailed derivations
             if self.config.solver_type == "midpoint":
                 x_t = (
                     (sigma_t / sigma_s0) * sample
@@ -859,7 +865,7 @@ class DPMSolverMultistepScheduler(SchedulerMixin, ConfigMixin):
                     + (alpha_t * ((torch.exp(-h) - 1.0) / h + 1.0)) * D1
                 )
         elif self.config.algorithm_type == "dpmsolver":
-            # See https://arxiv.org/abs/2206.00927 for detailed derivations
+            # See https://huggingface.co/papers/2206.00927 for detailed derivations
             if self.config.solver_type == "midpoint":
                 x_t = (
                     (alpha_t / alpha_s0) * sample
@@ -934,7 +940,7 @@ class DPMSolverMultistepScheduler(SchedulerMixin, ConfigMixin):
             if len(args) > 2:
                 sample = args[2]
             else:
-                raise ValueError(" missing`sample` as a required keyward argument")
+                raise ValueError("missing `sample` as a required keyword argument")
         if timestep_list is not None:
             deprecate(
                 "timestep_list",
@@ -975,7 +981,7 @@ class DPMSolverMultistepScheduler(SchedulerMixin, ConfigMixin):
         D1 = D1_0 + (r0 / (r0 + r1)) * (D1_0 - D1_1)
         D2 = (1.0 / (r0 + r1)) * (D1_0 - D1_1)
         if self.config.algorithm_type == "dpmsolver++":
-            # See https://arxiv.org/abs/2206.00927 for detailed derivations
+            # See https://huggingface.co/papers/2206.00927 for detailed derivations
             x_t = (
                 (sigma_t / sigma_s0) * sample
                 - (alpha_t * (torch.exp(-h) - 1.0)) * D0
@@ -983,7 +989,7 @@ class DPMSolverMultistepScheduler(SchedulerMixin, ConfigMixin):
                 - (alpha_t * ((torch.exp(-h) - 1.0 + h) / h**2 - 0.5)) * D2
             )
         elif self.config.algorithm_type == "dpmsolver":
-            # See https://arxiv.org/abs/2206.00927 for detailed derivations
+            # See https://huggingface.co/papers/2206.00927 for detailed derivations
             x_t = (
                 (alpha_t / alpha_s0) * sample
                 - (sigma_t * (torch.exp(h) - 1.0)) * D0

diffusers/schedulers/scheduling_dpmsolver_multistep_flax.py

@@ -1,4 +1,4 @@
-# Copyright 2024 TSAIL Team and The HuggingFace Team. All rights reserved.
+# Copyright 2025 TSAIL Team and The HuggingFace Team. All rights reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -80,14 +80,15 @@ class FlaxDPMSolverMultistepScheduler(FlaxSchedulerMixin, ConfigMixin):
     the convergence order guarantee. Empirically, sampling by DPM-Solver with only 20 steps can generate high-quality
     samples, and it can generate quite good samples even in only 10 steps.
 
-    For more details, see the original paper: https://arxiv.org/abs/2206.00927 and https://arxiv.org/abs/2211.01095
+    For more details, see the original paper: https://huggingface.co/papers/2206.00927 and
+    https://huggingface.co/papers/2211.01095
 
     Currently, we support the multistep DPM-Solver for both noise prediction models and data prediction models. We
     recommend to use `solver_order=2` for guided sampling, and `solver_order=3` for unconditional sampling.
 
-    We also support the "dynamic thresholding" method in Imagen (https://arxiv.org/abs/2205.11487). For pixel-space
-    diffusion models, you can set both `algorithm_type="dpmsolver++"` and `thresholding=True` to use the dynamic
-    thresholding. Note that the thresholding method is unsuitable for latent-space diffusion models (such as
+    We also support the "dynamic thresholding" method in Imagen (https://huggingface.co/papers/2205.11487). For
+    pixel-space diffusion models, you can set both `algorithm_type="dpmsolver++"` and `thresholding=True` to use the
+    dynamic thresholding. Note that the thresholding method is unsuitable for latent-space diffusion models (such as
     stable-diffusion).
 
     [`~ConfigMixin`] takes care of storing all config attributes that are passed in the scheduler's `__init__`
@@ -95,7 +96,8 @@ class FlaxDPMSolverMultistepScheduler(FlaxSchedulerMixin, ConfigMixin):
     [`SchedulerMixin`] provides general loading and saving functionality via the [`SchedulerMixin.save_pretrained`] and
     [`~SchedulerMixin.from_pretrained`] functions.
 
-    For more details, see the original paper: https://arxiv.org/abs/2206.00927 and https://arxiv.org/abs/2211.01095
+    For more details, see the original paper: https://huggingface.co/papers/2206.00927 and
+    https://huggingface.co/papers/2211.01095
 
     Args:
         num_train_timesteps (`int`): number of diffusion steps used to train the model.
@@ -113,21 +115,21 @@ class FlaxDPMSolverMultistepScheduler(FlaxSchedulerMixin, ConfigMixin):
             indicates whether the model predicts the noise (epsilon), or the data / `x0`. One of `epsilon`, `sample`,
             or `v-prediction`.
         thresholding (`bool`, default `False`):
-            whether to use the "dynamic thresholding" method (introduced by Imagen, https://arxiv.org/abs/2205.11487).
-            For pixel-space diffusion models, you can set both `algorithm_type=dpmsolver++` and `thresholding=True` to
-            use the dynamic thresholding. Note that the thresholding method is unsuitable for latent-space diffusion
-            models (such as stable-diffusion).
+            whether to use the "dynamic thresholding" method (introduced by Imagen,
+            https://huggingface.co/papers/2205.11487). For pixel-space diffusion models, you can set both
+            `algorithm_type=dpmsolver++` and `thresholding=True` to use the dynamic thresholding. Note that the
+            thresholding method is unsuitable for latent-space diffusion models (such as stable-diffusion).
         dynamic_thresholding_ratio (`float`, default `0.995`):
             the ratio for the dynamic thresholding method. Default is `0.995`, the same as Imagen
-            (https://arxiv.org/abs/2205.11487).
+            (https://huggingface.co/papers/2205.11487).
         sample_max_value (`float`, default `1.0`):
             the threshold value for dynamic thresholding. Valid only when `thresholding=True` and
             `algorithm_type="dpmsolver++`.
         algorithm_type (`str`, default `dpmsolver++`):
             the algorithm type for the solver. Either `dpmsolver` or `dpmsolver++`. The `dpmsolver` type implements the
-            algorithms in https://arxiv.org/abs/2206.00927, and the `dpmsolver++` type implements the algorithms in
-            https://arxiv.org/abs/2211.01095. We recommend to use `dpmsolver++` with `solver_order=2` for guided
-            sampling (e.g. stable-diffusion).
+            algorithms in https://huggingface.co/papers/2206.00927, and the `dpmsolver++` type implements the
+            algorithms in https://huggingface.co/papers/2211.01095. We recommend to use `dpmsolver++` with
+            `solver_order=2` for guided sampling (e.g. stable-diffusion).
         solver_type (`str`, default `midpoint`):
             the solver type for the second-order solver. Either `midpoint` or `heun`. The solver type slightly affects
             the sample quality, especially for small number of steps. We empirically find that `midpoint` solvers are
@@ -297,7 +299,7 @@ class FlaxDPMSolverMultistepScheduler(FlaxSchedulerMixin, ConfigMixin):
                 )
 
             if self.config.thresholding:
-                # Dynamic thresholding in https://arxiv.org/abs/2205.11487
+                # Dynamic thresholding in https://huggingface.co/papers/2205.11487
                 dynamic_max_val = jnp.percentile(
                     jnp.abs(x0_pred), self.config.dynamic_thresholding_ratio, axis=tuple(range(1, x0_pred.ndim))
                 )
@@ -335,7 +337,7 @@ class FlaxDPMSolverMultistepScheduler(FlaxSchedulerMixin, ConfigMixin):
         """
         One step for the first-order DPM-Solver (equivalent to DDIM).
 
-        See https://arxiv.org/abs/2206.00927 for the detailed derivation.
+        See https://huggingface.co/papers/2206.00927 for the detailed derivation.
 
         Args:
             model_output (`jnp.ndarray`): direct output from learned diffusion model.
@@ -390,7 +392,7 @@ class FlaxDPMSolverMultistepScheduler(FlaxSchedulerMixin, ConfigMixin):
         r0 = h_0 / h
         D0, D1 = m0, (1.0 / r0) * (m0 - m1)
         if self.config.algorithm_type == "dpmsolver++":
-            # See https://arxiv.org/abs/2211.01095 for detailed derivations
+            # See https://huggingface.co/papers/2211.01095 for detailed derivations
             if self.config.solver_type == "midpoint":
                 x_t = (
                     (sigma_t / sigma_s0) * sample
@@ -404,7 +406,7 @@ class FlaxDPMSolverMultistepScheduler(FlaxSchedulerMixin, ConfigMixin):
                     + (alpha_t * ((jnp.exp(-h) - 1.0) / h + 1.0)) * D1
                 )
         elif self.config.algorithm_type == "dpmsolver":
-            # See https://arxiv.org/abs/2206.00927 for detailed derivations
+            # See https://huggingface.co/papers/2206.00927 for detailed derivations
             if self.config.solver_type == "midpoint":
                 x_t = (
                     (alpha_t / alpha_s0) * sample
@@ -458,7 +460,7 @@ class FlaxDPMSolverMultistepScheduler(FlaxSchedulerMixin, ConfigMixin):
         D1 = D1_0 + (r0 / (r0 + r1)) * (D1_0 - D1_1)
         D2 = (1.0 / (r0 + r1)) * (D1_0 - D1_1)
         if self.config.algorithm_type == "dpmsolver++":
-            # See https://arxiv.org/abs/2206.00927 for detailed derivations
+            # See https://huggingface.co/papers/2206.00927 for detailed derivations
             x_t = (
                 (sigma_t / sigma_s0) * sample
                 - (alpha_t * (jnp.exp(-h) - 1.0)) * D0
@@ -466,7 +468,7 @@ class FlaxDPMSolverMultistepScheduler(FlaxSchedulerMixin, ConfigMixin):
                 - (alpha_t * ((jnp.exp(-h) - 1.0 + h) / h**2 - 0.5)) * D2
             )
         elif self.config.algorithm_type == "dpmsolver":
-            # See https://arxiv.org/abs/2206.00927 for detailed derivations
+            # See https://huggingface.co/papers/2206.00927 for detailed derivations
             x_t = (
                 (alpha_t / alpha_s0) * sample
                 - (sigma_t * (jnp.exp(h) - 1.0)) * D0

diffusers/schedulers/scheduling_dpmsolver_multistep_inverse.py

@@ -1,4 +1,4 @@
-# Copyright 2024 TSAIL Team and The HuggingFace Team. All rights reserved.
+# Copyright 2025 TSAIL Team and The HuggingFace Team. All rights reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -257,7 +257,7 @@ class DPMSolverMultistepInverseScheduler(SchedulerMixin, ConfigMixin):
         clipped_idx = torch.searchsorted(torch.flip(self.lambda_t, [0]), self.config.lambda_min_clipped).item()
         self.noisiest_timestep = self.config.num_train_timesteps - 1 - clipped_idx
 
-        # "linspace", "leading", "trailing" corresponds to annotation of Table 2. of https://arxiv.org/abs/2305.08891
+        # "linspace", "leading", "trailing" corresponds to annotation of Table 2. of https://huggingface.co/papers/2305.08891
         if self.config.timestep_spacing == "linspace":
             timesteps = (
                 np.linspace(0, self.noisiest_timestep, num_inference_steps + 1).round()[:-1].copy().astype(np.int64)
@@ -338,7 +338,7 @@ class DPMSolverMultistepInverseScheduler(SchedulerMixin, ConfigMixin):
         pixels from saturation at each step. We find that dynamic thresholding results in significantly better
         photorealism as well as better image-text alignment, especially when using very large guidance weights."
 
-        https://arxiv.org/abs/2205.11487
+        https://huggingface.co/papers/2205.11487
         """
         dtype = sample.dtype
         batch_size, channels, *remaining_dims = sample.shape
@@ -513,7 +513,7 @@ class DPMSolverMultistepInverseScheduler(SchedulerMixin, ConfigMixin):
             if len(args) > 1:
                 sample = args[1]
             else:
-                raise ValueError("missing `sample` as a required keyward argument")
+                raise ValueError("missing `sample` as a required keyword argument")
         if timestep is not None:
             deprecate(
                 "timesteps",
@@ -609,7 +609,7 @@ class DPMSolverMultistepInverseScheduler(SchedulerMixin, ConfigMixin):
             if len(args) > 2:
                 sample = args[2]
             else:
-                raise ValueError(" missing `sample` as a required keyward argument")
+                raise ValueError("missing `sample` as a required keyword argument")
         if timestep is not None:
             deprecate(
                 "timesteps",
@@ -679,7 +679,7 @@ class DPMSolverMultistepInverseScheduler(SchedulerMixin, ConfigMixin):
             if len(args) > 2:
                 sample = args[2]
             else:
-                raise ValueError(" missing `sample` as a required keyward argument")
+                raise ValueError("missing `sample` as a required keyword argument")
         if timestep_list is not None:
             deprecate(
                 "timestep_list",
@@ -714,7 +714,7 @@ class DPMSolverMultistepInverseScheduler(SchedulerMixin, ConfigMixin):
         r0 = h_0 / h
         D0, D1 = m0, (1.0 / r0) * (m0 - m1)
         if self.config.algorithm_type == "dpmsolver++":
-            # See https://arxiv.org/abs/2211.01095 for detailed derivations
+            # See https://huggingface.co/papers/2211.01095 for detailed derivations
             if self.config.solver_type == "midpoint":
                 x_t = (
                     (sigma_t / sigma_s0) * sample
@@ -728,7 +728,7 @@ class DPMSolverMultistepInverseScheduler(SchedulerMixin, ConfigMixin):
                     + (alpha_t * ((torch.exp(-h) - 1.0) / h + 1.0)) * D1
                 )
         elif self.config.algorithm_type == "dpmsolver":
-            # See https://arxiv.org/abs/2206.00927 for detailed derivations
+            # See https://huggingface.co/papers/2206.00927 for detailed derivations
             if self.config.solver_type == "midpoint":
                 x_t = (
                     (alpha_t / alpha_s0) * sample
@@ -804,7 +804,7 @@ class DPMSolverMultistepInverseScheduler(SchedulerMixin, ConfigMixin):
             if len(args) > 2:
                 sample = args[2]
             else:
-                raise ValueError(" missing`sample` as a required keyward argument")
+                raise ValueError("missing `sample` as a required keyword argument")
         if timestep_list is not None:
             deprecate(
                 "timestep_list",
@@ -845,7 +845,7 @@ class DPMSolverMultistepInverseScheduler(SchedulerMixin, ConfigMixin):
         D1 = D1_0 + (r0 / (r0 + r1)) * (D1_0 - D1_1)
         D2 = (1.0 / (r0 + r1)) * (D1_0 - D1_1)
         if self.config.algorithm_type == "dpmsolver++":
-            # See https://arxiv.org/abs/2206.00927 for detailed derivations
+            # See https://huggingface.co/papers/2206.00927 for detailed derivations
             x_t = (
                 (sigma_t / sigma_s0) * sample
                 - (alpha_t * (torch.exp(-h) - 1.0)) * D0
@@ -853,7 +853,7 @@ class DPMSolverMultistepInverseScheduler(SchedulerMixin, ConfigMixin):
                 - (alpha_t * ((torch.exp(-h) - 1.0 + h) / h**2 - 0.5)) * D2
             )
         elif self.config.algorithm_type == "dpmsolver":
-            # See https://arxiv.org/abs/2206.00927 for detailed derivations
+            # See https://huggingface.co/papers/2206.00927 for detailed derivations
             x_t = (
                 (alpha_t / alpha_s0) * sample
                 - (sigma_t * (torch.exp(h) - 1.0)) * D0

diffusers/schedulers/scheduling_dpmsolver_sde.py

@@ -1,4 +1,4 @@
-# Copyright 2024 Katherine Crowson, The HuggingFace Team and hlky. All rights reserved.
+# Copyright 2025 Katherine Crowson, The HuggingFace Team and hlky. All rights reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -352,7 +352,7 @@ class DPMSolverSDEScheduler(SchedulerMixin, ConfigMixin):
 
         num_train_timesteps = num_train_timesteps or self.config.num_train_timesteps
 
-        # "linspace", "leading", "trailing" corresponds to annotation of Table 2. of https://arxiv.org/abs/2305.08891
+        # "linspace", "leading", "trailing" corresponds to annotation of Table 2. of https://huggingface.co/papers/2305.08891
         if self.config.timestep_spacing == "linspace":
             timesteps = np.linspace(0, num_train_timesteps - 1, num_inference_steps, dtype=float)[::-1].copy()
         elif self.config.timestep_spacing == "leading":

diffusers/schedulers/scheduling_dpmsolver_singlestep.py

@@ -1,4 +1,4 @@
-# Copyright 2024 TSAIL Team and The HuggingFace Team. All rights reserved.
+# Copyright 2025 TSAIL Team and The HuggingFace Team. All rights reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -169,6 +169,8 @@ class DPMSolverSinglestepScheduler(SchedulerMixin, ConfigMixin):
         final_sigmas_type: Optional[str] = "zero",  # "zero", "sigma_min"
         lambda_min_clipped: float = -float("inf"),
         variance_type: Optional[str] = None,
+        use_dynamic_shifting: bool = False,
+        time_shift_type: str = "exponential",
     ):
         if self.config.use_beta_sigmas and not is_scipy_available():
             raise ImportError("Make sure to install scipy if you want to use beta sigmas.")
@@ -218,7 +220,7 @@ class DPMSolverSinglestepScheduler(SchedulerMixin, ConfigMixin):
 
         if algorithm_type not in ["dpmsolver++", "sde-dpmsolver++"] and final_sigmas_type == "zero":
             raise ValueError(
-                f"`final_sigmas_type` {final_sigmas_type} is not supported for `algorithm_type` {algorithm_type}. Please chooose `sigma_min` instead."
+                f"`final_sigmas_type` {final_sigmas_type} is not supported for `algorithm_type` {algorithm_type}. Please choose `sigma_min` instead."
             )
 
         # setable values
@@ -301,6 +303,7 @@ class DPMSolverSinglestepScheduler(SchedulerMixin, ConfigMixin):
         self,
         num_inference_steps: int = None,
         device: Union[str, torch.device] = None,
+        mu: Optional[float] = None,
         timesteps: Optional[List[int]] = None,
     ):
         """
@@ -316,6 +319,9 @@ class DPMSolverSinglestepScheduler(SchedulerMixin, ConfigMixin):
                 timestep spacing strategy of equal spacing between timesteps schedule is used. If `timesteps` is
                 passed, `num_inference_steps` must be `None`.
         """
+        if mu is not None:
+            assert self.config.use_dynamic_shifting and self.config.time_shift_type == "exponential"
+            self.config.flow_shift = np.exp(mu)
         if num_inference_steps is None and timesteps is None:
             raise ValueError("Must pass exactly one of  `num_inference_steps` or `timesteps`.")
         if num_inference_steps is not None and timesteps is not None:
@@ -410,7 +416,7 @@ class DPMSolverSinglestepScheduler(SchedulerMixin, ConfigMixin):
         pixels from saturation at each step. We find that dynamic thresholding results in significantly better
         photorealism as well as better image-text alignment, especially when using very large guidance weights."
 
-        https://arxiv.org/abs/2205.11487
+        https://huggingface.co/papers/2205.11487
         """
         dtype = sample.dtype
         batch_size, channels, *remaining_dims = sample.shape
@@ -584,7 +590,7 @@ class DPMSolverSinglestepScheduler(SchedulerMixin, ConfigMixin):
             if len(args) > 1:
                 sample = args[1]
             else:
-                raise ValueError("missing `sample` as a required keyward argument")
+                raise ValueError("missing `sample` as a required keyword argument")
         if timestep is not None:
             deprecate(
                 "timesteps",
@@ -681,7 +687,7 @@ class DPMSolverSinglestepScheduler(SchedulerMixin, ConfigMixin):
             if len(args) > 2:
                 sample = args[2]
             else:
-                raise ValueError(" missing `sample` as a required keyward argument")
+                raise ValueError("missing `sample` as a required keyword argument")
         if timestep is not None:
             deprecate(
                 "timesteps",
@@ -746,7 +752,7 @@ class DPMSolverSinglestepScheduler(SchedulerMixin, ConfigMixin):
             if len(args) > 2:
                 sample = args[2]
             else:
-                raise ValueError(" missing `sample` as a required keyward argument")
+                raise ValueError("missing `sample` as a required keyword argument")
         if timestep_list is not None:
             deprecate(
                 "timestep_list",
@@ -780,7 +786,7 @@ class DPMSolverSinglestepScheduler(SchedulerMixin, ConfigMixin):
         r0 = h_0 / h
         D0, D1 = m1, (1.0 / r0) * (m0 - m1)
         if self.config.algorithm_type == "dpmsolver++":
-            # See https://arxiv.org/abs/2211.01095 for detailed derivations
+            # See https://huggingface.co/papers/2211.01095 for detailed derivations
             if self.config.solver_type == "midpoint":
                 x_t = (
                     (sigma_t / sigma_s1) * sample
@@ -794,7 +800,7 @@ class DPMSolverSinglestepScheduler(SchedulerMixin, ConfigMixin):
                     + (alpha_t * ((torch.exp(-h) - 1.0) / h + 1.0)) * D1
                 )
         elif self.config.algorithm_type == "dpmsolver":
-            # See https://arxiv.org/abs/2206.00927 for detailed derivations
+            # See https://huggingface.co/papers/2206.00927 for detailed derivations
             if self.config.solver_type == "midpoint":
                 x_t = (
                     (alpha_t / alpha_s1) * sample
@@ -858,7 +864,7 @@ class DPMSolverSinglestepScheduler(SchedulerMixin, ConfigMixin):
             if len(args) > 2:
                 sample = args[2]
             else:
-                raise ValueError(" missing`sample` as a required keyward argument")
+                raise ValueError("missing `sample` as a required keyword argument")
         if timestep_list is not None:
             deprecate(
                 "timestep_list",
@@ -899,7 +905,7 @@ class DPMSolverSinglestepScheduler(SchedulerMixin, ConfigMixin):
         D1 = (r0 * D1_0 - r1 * D1_1) / (r0 - r1)
         D2 = 2.0 * (D1_1 - D1_0) / (r0 - r1)
         if self.config.algorithm_type == "dpmsolver++":
-            # See https://arxiv.org/abs/2206.00927 for detailed derivations
+            # See https://huggingface.co/papers/2206.00927 for detailed derivations
             if self.config.solver_type == "midpoint":
                 x_t = (
                     (sigma_t / sigma_s2) * sample
@@ -914,7 +920,7 @@ class DPMSolverSinglestepScheduler(SchedulerMixin, ConfigMixin):
                     - (alpha_t * ((torch.exp(-h) - 1.0 + h) / h**2 - 0.5)) * D2
                 )
         elif self.config.algorithm_type == "dpmsolver":
-            # See https://arxiv.org/abs/2206.00927 for detailed derivations
+            # See https://huggingface.co/papers/2206.00927 for detailed derivations
             if self.config.solver_type == "midpoint":
                 x_t = (
                     (alpha_t / alpha_s2) * sample
@@ -981,12 +987,12 @@ class DPMSolverSinglestepScheduler(SchedulerMixin, ConfigMixin):
             if len(args) > 2:
                 sample = args[2]
             else:
-                raise ValueError(" missing`sample` as a required keyward argument")
+                raise ValueError("missing `sample` as a required keyword argument")
         if order is None:
             if len(args) > 3:
                 order = args[3]
             else:
-                raise ValueError(" missing `order` as a required keyward argument")
+                raise ValueError("missing `order` as a required keyword argument")
         if timestep_list is not None:
             deprecate(
                 "timestep_list",

diffusers/schedulers/scheduling_edm_dpmsolver_multistep.py

@@ -1,4 +1,4 @@
-# Copyright 2024 TSAIL Team and The HuggingFace Team. All rights reserved.
+# Copyright 2025 TSAIL Team and The HuggingFace Team. All rights reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -31,7 +31,7 @@ class EDMDPMSolverMultistepScheduler(SchedulerMixin, ConfigMixin):
     `EDMDPMSolverMultistepScheduler` is a fast dedicated high-order solver for diffusion ODEs.
 
     [1] Karras, Tero, et al. "Elucidating the Design Space of Diffusion-Based Generative Models."
-    https://arxiv.org/abs/2206.00364
+    https://huggingface.co/papers/2206.00364
 
     This model inherits from [`SchedulerMixin`] and [`ConfigMixin`]. Check the superclass documentation for the generic
     methods the library implements for all schedulers such as loading and saving.
@@ -47,8 +47,8 @@ class EDMDPMSolverMultistepScheduler(SchedulerMixin, ConfigMixin):
             The standard deviation of the data distribution. This is set to 0.5 in the EDM paper [1].
         sigma_schedule (`str`, *optional*, defaults to `karras`):
             Sigma schedule to compute the `sigmas`. By default, we the schedule introduced in the EDM paper
-            (https://arxiv.org/abs/2206.00364). Other acceptable value is "exponential". The exponential schedule was
-            incorporated in this model: https://huggingface.co/stabilityai/cosxl.
+            (https://huggingface.co/papers/2206.00364). Other acceptable value is "exponential". The exponential
+            schedule was incorporated in this model: https://huggingface.co/stabilityai/cosxl.
         num_train_timesteps (`int`, defaults to 1000):
             The number of diffusion steps to train the model.
         solver_order (`int`, defaults to 2):
@@ -176,7 +176,7 @@ class EDMDPMSolverMultistepScheduler(SchedulerMixin, ConfigMixin):
 
     # Copied from diffusers.schedulers.scheduling_edm_euler.EDMEulerScheduler.precondition_inputs
     def precondition_inputs(self, sample, sigma):
-        c_in = 1 / ((sigma**2 + self.config.sigma_data**2) ** 0.5)
+        c_in = self._get_conditioning_c_in(sigma)
         scaled_sample = sample * c_in
         return scaled_sample
 
@@ -305,7 +305,7 @@ class EDMDPMSolverMultistepScheduler(SchedulerMixin, ConfigMixin):
         pixels from saturation at each step. We find that dynamic thresholding results in significantly better
         photorealism as well as better image-text alignment, especially when using very large guidance weights."
 
-        https://arxiv.org/abs/2205.11487
+        https://huggingface.co/papers/2205.11487
         """
         dtype = sample.dtype
         batch_size, channels, *remaining_dims = sample.shape
@@ -472,7 +472,7 @@ class EDMDPMSolverMultistepScheduler(SchedulerMixin, ConfigMixin):
         r0 = h_0 / h
         D0, D1 = m0, (1.0 / r0) * (m0 - m1)
         if self.config.algorithm_type == "dpmsolver++":
-            # See https://arxiv.org/abs/2211.01095 for detailed derivations
+            # See https://huggingface.co/papers/2211.01095 for detailed derivations
             if self.config.solver_type == "midpoint":
                 x_t = (
                     (sigma_t / sigma_s0) * sample
@@ -548,7 +548,7 @@ class EDMDPMSolverMultistepScheduler(SchedulerMixin, ConfigMixin):
         D1 = D1_0 + (r0 / (r0 + r1)) * (D1_0 - D1_1)
         D2 = (1.0 / (r0 + r1)) * (D1_0 - D1_1)
         if self.config.algorithm_type == "dpmsolver++":
-            # See https://arxiv.org/abs/2206.00927 for detailed derivations
+            # See https://huggingface.co/papers/2206.00927 for detailed derivations
             x_t = (
                 (sigma_t / sigma_s0) * sample
                 - (alpha_t * (torch.exp(-h) - 1.0)) * D0
@@ -703,5 +703,10 @@ class EDMDPMSolverMultistepScheduler(SchedulerMixin, ConfigMixin):
         noisy_samples = original_samples + noise * sigma
         return noisy_samples
 
+    # Copied from diffusers.schedulers.scheduling_edm_euler.EDMEulerScheduler._get_conditioning_c_in
+    def _get_conditioning_c_in(self, sigma):
+        c_in = 1 / ((sigma**2 + self.config.sigma_data**2) ** 0.5)
+        return c_in
+
     def __len__(self):
         return self.config.num_train_timesteps