with RTX 5090 200GB RAM - ltx 2.3 is broken; it runs out of memory (OOM) when I increase the frame count, but even when I drop it to 120 frames, it still OOMs—despite working perfectly before. I’ve tested it on multiple instances across VAST and RunPod

WORKFLOW: Job ceb6b93e-32c1-49e8-8232-406eb2c1b9a6.json

from large pool | 0 MiB | 0 MiB | 0 B | 0 B |

| from small pool | 0 MiB | 0 MiB | 0 B | 0 B |

|---------------------------------------------------------------------------|

| Active memory | 29878 MiB | 30453 MiB | 0 B | 0 B |

| from large pool | 0 MiB | 0 MiB | 0 B | 0 B |

| from small pool | 0 MiB | 0 MiB | 0 B | 0 B |

|---------------------------------------------------------------------------|

| Requested memory | 0 B | 0 B | 0 B | 0 B |

| from large pool | 0 B | 0 B | 0 B | 0 B |

| from small pool | 0 B | 0 B | 0 B | 0 B |

|---------------------------------------------------------------------------|

| GPU reserved memory | 30880 MiB | 30880 MiB | 0 B | 0 B |

| from large pool | 0 MiB | 0 MiB | 0 B | 0 B |

| from small pool | 0 MiB | 0 MiB | 0 B | 0 B |

|---------------------------------------------------------------------------|

| Non-releasable memory | 0 B | 0 B | 0 B | 0 B |

| from large pool | 0 B | 0 B | 0 B | 0 B |

| from small pool | 0 B | 0 B | 0 B | 0 B |

|---------------------------------------------------------------------------|

| Allocations | 0 | 0 | 0 | 0 |

| from large pool | 0 | 0 | 0 | 0 |

| from small pool | 0 | 0 | 0 | 0 |

|---------------------------------------------------------------------------|

| Active allocs | 0 | 0 | 0 | 0 |

| from large pool | 0 | 0 | 0 | 0 |

| from small pool | 0 | 0 | 0 | 0 |

|---------------------------------------------------------------------------|

| GPU reserved segments | 0 | 0 | 0 | 0 |

| from large pool | 0 | 0 | 0 | 0 |

| from small pool | 0 | 0 | 0 | 0 |

|---------------------------------------------------------------------------|

| Non-releasable allocs | 0 | 0 | 0 | 0 |

| from large pool | 0 | 0 | 0 | 0 |

| from small pool | 0 | 0 | 0 | 0 |

|---------------------------------------------------------------------------|

| Oversize allocations | 0 | 0 | 0 | 0 |

|---------------------------------------------------------------------------|

| Oversize GPU segments | 0 | 0 | 0 | 0 |

|===========================================================================|

[ERROR] Got an OOM, unloading all loaded models.

[INFO] Prompt executed in 70.11 seconds

[INFO] Requested to load LTXAV

[INFO] loaded partially; 27620.41 MB usable, 27580.40 MB loaded, 1208.96
MB offloaded, 40.01 MB buffer reserved, lowvram patches: 407

[INFO] Patching torch settings: torch.backends.cuda.matmul.allow_fp16_accumulation = True

33%|████████████████████████████████████████████████████████▎

| 1/3 [00:04<00:09, 4.58s/it]

[INFO] Patching torch settings: torch.backends.cuda.matmul.allow_fp16_accumulation = False

→ [10S] TiledSampler v2.5: carrier wrapper sampling failed
(OutOfMemoryError: Allocation on device 0 would exceed allowed memory.
(out of memory)

Currently allocated : 29.60 GiB

Requested : 128.91 MiB

Device limit : 31.36 GiB

Free (according to CUDA): 28.69 MiB

PyTorch limit (set by user-supplied memory fraction)

: 17179869184.00 GiB); falling back to plain video for carrier tile

[INFO] 0 models unloaded.

[INFO] Unloaded partially: 635.17 MB freed, 26945.23 MB remains loaded, 40.01 MB buffer reserved, lowvram patches: 787

[INFO] Patching torch settings: torch.backends.cuda.matmul.allow_fp16_accumulation = True

0%|

| 0/3 [00:00<?, ?it/s]

[INFO] Patching torch settings: torch.backends.cuda.matmul.allow_fp16_accumulation = False

[ERROR] !!! Exception during processing !!! Allocation on device 0 would exceed allowed memory. (out of memory)

Currently allocated : 29.51 GiB

Requested : 550.00 MiB

Device limit : 31.36 GiB

Free (according to CUDA): 28.69 MiB

PyTorch limit (set by user-supplied memory fraction)

: 17179869184.00 GiB

[ERROR] Traceback (most recent call last):

File "/workspace/ComfyUI/custom_nodes/10S-Comfy-nodes/latent_tiled_sampler.py", line 923, in sample_tiled

tile_result = guider.sample(

^^^^^^^^^^^^^^

File "/workspace/ComfyUI/comfy/samplers.py", line 1316, in sample

output = executor.execute(noise, latent_image, sampler, sigmas,
denoise_mask, callback, disable_pbar, seed, latent_shapes=latent_shapes)

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/workspace/ComfyUI/comfy/patcher_extension.py", line 113, in execute

return self.original(*args, **kwargs)

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/workspace/ComfyUI/comfy/samplers.py", line 1254, in outer_sample

output = self.inner_sample(noise, latent_image, device, sampler, sigmas,
denoise_mask, callback, disable_pbar, seed,
latent_shapes=latent_shapes)

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/workspace/ComfyUI/comfy/samplers.py", line 1229, in inner_sample

samples = executor.execute(self, sigmas, extra_args, callback, noise, latent_image, denoise_mask, disable_pbar)

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/workspace/ComfyUI/comfy/patcher_extension.py", line 113, in execute

return self.original(*args, **kwargs)

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/workspace/ComfyUI/comfy/samplers.py", line 999, in sample

samples = self.sampler_function(model_k, noise, sigmas,
extra_args=extra_args, callback=k_callback, disable=disable_pbar,
**self.extra_options)

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/venv/main/lib/python3.12/site-packages/torch/utils/_contextlib.py", line 124, in decorate_context

return func(*args, **kwargs)

^^^^^^^^^^^^^^^^^^^^^

File "/workspace/ComfyUI/comfy/k_diffusion/sampling.py", line 1288, in sample_euler_ancestral_cfg_pp

denoised = model(x, sigmas[i] * s_in, **extra_args)

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/workspace/ComfyUI/comfy/samplers.py", line 639, in call

out = self.inner_model(x, sigma, model_options=model_options, seed=seed)

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/workspace/ComfyUI/comfy/samplers.py", line 1202, in call

return self.outer_predict_noise(*args, **kwargs)

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/workspace/ComfyUI/comfy/samplers.py", line 1209, in outer_predict_noise

).execute(x, timestep, model_options, seed)

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/workspace/ComfyUI/comfy/patcher_extension.py", line 113, in execute

return self.original(*args, **kwargs)

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/workspace/ComfyUI/comfy/samplers.py", line 1212, in predict_noise

return sampling_function(self.inner_model, x, timestep,
self.conds.get("negative", None), self.conds.get("positive", None),
self.cfg, model_options=model_options, seed=seed)

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/workspace/ComfyUI/comfy/samplers.py", line 619, in sampling_function

out = calc_cond_batch(model, conds, x, timestep, model_options)

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/workspace/ComfyUI/comfy/samplers.py", line 210, in calc_cond_batch

return _calc_cond_batch_outer(model, conds, x_in, timestep, model_options)

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/workspace/ComfyUI/comfy/samplers.py", line 218, in _calc_cond_batch_outer

return executor.execute(model, conds, x_in, timestep, model_options)

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/workspace/ComfyUI/comfy/patcher_extension.py", line 113, in execute

return self.original(*args, **kwargs)

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/workspace/ComfyUI/comfy/samplers.py", line 334, in calc_cond_batch

output = model.apply_model(input_x, timestep, **c).chunk(batch_chunks)

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/workspace/ComfyUI/comfy/model_base.py", line 191, in apply_model

return comfy.patcher_extension.WrapperExecutor.new_class_executor(

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/workspace/ComfyUI/comfy/patcher_extension.py", line 113, in execute

return self.original(*args, **kwargs)

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/workspace/ComfyUI/comfy/model_base.py", line 235, in _apply_model

model_output = self.diffusion_model(xc, t, context=context,
control=control, transformer_options=transformer_options, **extra_conds)

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/venv/main/lib/python3.12/site-packages/torch/nn/modules/module.py", line 1779, in _wrapped_call_impl

return self._call_impl(*args, **kwargs)

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/venv/main/lib/python3.12/site-packages/torch/nn/modules/module.py", line 1885, in _call_impl

return inner()

^^^^^^^

File "/venv/main/lib/python3.12/site-packages/torch/nn/modules/module.py", line 1833, in inner

result = forward_call(*args, **kwargs)

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/workspace/ComfyUI/comfy/ldm/lightricks/av_model.py", line 1069, in forward

return super().forward(

^^^^^^^^^^^^^^^^

File "/workspace/ComfyUI/comfy/ldm/lightricks/model.py", line 936, in forward

return comfy.patcher_extension.WrapperExecutor.new_class_executor(

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/workspace/ComfyUI/comfy/patcher_extension.py", line 113, in execute

return self.original(*args, **kwargs)

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/workspace/ComfyUI/comfy/ldm/lightricks/model.py", line 989, in _forward

x = self._process_transformer_blocks(

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/workspace/ComfyUI/comfy/ldm/lightricks/av_model.py", line 969, in _process_transformer_blocks

vx, ax = block(

^^^^^^

File "/venv/main/lib/python3.12/site-packages/torch/nn/modules/module.py", line 1779, in _wrapped_call_impl

return self._call_impl(*args, **kwargs)

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/venv/main/lib/python3.12/site-packages/torch/nn/modules/module.py", line 1790, in _call_impl

return forward_call(*args, **kwargs)

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/workspace/ComfyUI/comfy/ldm/lightricks/av_model.py", line 276, in forward

attn1_out = self.attn1(norm_vx, pe=v_pe, mask=self_attention_mask, transformer_options=transformer_options)

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/venv/main/lib/python3.12/site-packages/torch/nn/modules/module.py", line 1779, in _wrapped_call_impl

return self._call_impl(*args, **kwargs)

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/venv/main/lib/python3.12/site-packages/torch/nn/modules/module.py", line 1885, in _call_impl

return inner()

^^^^^^^

File "/venv/main/lib/python3.12/site-packages/torch/nn/modules/module.py", line 1846, in inner

hook_result = hook(self, args, result)

^^^^^^^^^^^^^^^^^^^^^^^^

File "/workspace/ComfyUI/custom_nodes/10S-Comfy-nodes/latent_likeness_anchor.py", line 744, in hook

pull = pull * frame_weights.view(1, F_gen_count, 1, 1)

~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

torch.OutOfMemoryError: Allocation on device 0 would exceed allowed memory. (out of memory)

Currently allocated : 29.60 GiB

Requested : 128.91 MiB

Device limit : 31.36 GiB

Free (according to CUDA): 28.69 MiB

PyTorch limit (set by user-supplied memory fraction)

: 17179869184.00 GiB

During handling of the above exception, another exception occurred:

Traceback (most recent call last):

File "/workspace/ComfyUI/execution.py", line 542, in execute

output_data, output_ui, has_subgraph, has_pending_tasks = await
get_output_data(prompt_id, unique_id, obj, input_data_all,
execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb,
v3_data=v3_data)

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/workspace/ComfyUI/execution.py", line 341, in get_output_data

return_values = await _async_map_node_over_list(prompt_id, unique_id,
obj, input_data_all, obj.FUNCTION, allow_interrupt=True,
execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb,
v3_data=v3_data)

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/workspace/ComfyUI/execution.py", line 315, in _async_map_node_over_list

await process_inputs(input_dict, i)

File "/workspace/ComfyUI/execution.py", line 303, in process_inputs

result = f(**inputs)

^^^^^^^^^^^

File "/workspace/ComfyUI/custom_nodes/10S-Comfy-nodes/latent_tiled_sampler.py", line 1078, in sample_tiled

tile_samples = guider.sample(

^^^^^^^^^^^^^^

File "/workspace/ComfyUI/comfy/samplers.py", line 1316, in sample

output = executor.execute(noise, latent_image, sampler, sigmas,
denoise_mask, callback, disable_pbar, seed, latent_shapes=latent_shapes)

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/workspace/ComfyUI/comfy/patcher_extension.py", line 113, in execute

return self.original(*args, **kwargs)

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/workspace/ComfyUI/comfy/samplers.py", line 1254, in outer_sample

output = self.inner_sample(noise, latent_image, device, sampler, sigmas,
denoise_mask, callback, disable_pbar, seed,
latent_shapes=latent_shapes)

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/workspace/ComfyUI/comfy/samplers.py", line 1229, in inner_sample

samples = executor.execute(self, sigmas, extra_args, callback, noise, latent_image, denoise_mask, disable_pbar)

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/workspace/ComfyUI/comfy/patcher_extension.py", line 113, in execute

return self.original(*args, **kwargs)

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/workspace/ComfyUI/comfy/samplers.py", line 999, in sample

samples = self.sampler_function(model_k, noise, sigmas,
extra_args=extra_args, callback=k_callback, disable=disable_pbar,
**self.extra_options)

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/venv/main/lib/python3.12/site-packages/torch/utils/_contextlib.py", line 124, in decorate_context

return func(*args, **kwargs)

^^^^^^^^^^^^^^^^^^^^^

File "/workspace/ComfyUI/comfy/k_diffusion/sampling.py", line 1288, in sample_euler_ancestral_cfg_pp

denoised = model(x, sigmas[i] * s_in, **extra_args)

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/workspace/ComfyUI/comfy/samplers.py", line 639, in call

out = self.inner_model(x, sigma, model_options=model_options, seed=seed)

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/workspace/ComfyUI/comfy/samplers.py", line 1202, in call

return self.outer_predict_noise(*args, **kwargs)

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/workspace/ComfyUI/comfy/samplers.py", line 1209, in outer_predict_noise

).execute(x, timestep, model_options, seed)

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/workspace/ComfyUI/comfy/patcher_extension.py", line 113, in execute

return self.original(*args, **kwargs)

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/workspace/ComfyUI/comfy/samplers.py", line 1212, in predict_noise

return sampling_function(self.inner_model, x, timestep,
self.conds.get("negative", None), self.conds.get("positive", None),
self.cfg, model_options=model_options, seed=seed)

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/workspace/ComfyUI/comfy/samplers.py", line 619, in sampling_function

out = calc_cond_batch(model, conds, x, timestep, model_options)

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/workspace/ComfyUI/comfy/samplers.py", line 210, in calc_cond_batch

return _calc_cond_batch_outer(model, conds, x_in, timestep, model_options)

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/workspace/ComfyUI/comfy/samplers.py", line 218, in _calc_cond_batch_outer

return executor.execute(model, conds, x_in, timestep, model_options)

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/workspace/ComfyUI/comfy/patcher_extension.py", line 113, in execute

return self.original(*args, **kwargs)

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/workspace/ComfyUI/comfy/samplers.py", line 334, in calc_cond_batch

output = model.apply_model(input_x, timestep, **c).chunk(batch_chunks)

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/workspace/ComfyUI/comfy/model_base.py", line 191, in apply_model

return comfy.patcher_extension.WrapperExecutor.new_class_executor(

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/workspace/ComfyUI/comfy/patcher_extension.py", line 113, in execute

return self.original(*args, **kwargs)

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/workspace/ComfyUI/comfy/model_base.py", line 235, in _apply_model

model_output = self.diffusion_model(xc, t, context=context,
control=control, transformer_options=transformer_options, **extra_conds)

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/venv/main/lib/python3.12/site-packages/torch/nn/modules/module.py", line 1779, in _wrapped_call_impl

return self._call_impl(*args, **kwargs)

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/venv/main/lib/python3.12/site-packages/torch/nn/modules/module.py", line 1885, in _call_impl

return inner()

^^^^^^^

File "/venv/main/lib/python3.12/site-packages/torch/nn/modules/module.py", line 1833, in inner

result = forward_call(*args, **kwargs)

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/workspace/ComfyUI/comfy/ldm/lightricks/av_model.py", line 1069, in forward

return super().forward(

^^^^^^^^^^^^^^^^

File "/workspace/ComfyUI/comfy/ldm/lightricks/model.py", line 936, in forward

return comfy.patcher_extension.WrapperExecutor.new_class_executor(

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/workspace/ComfyUI/comfy/patcher_extension.py", line 113, in execute

return self.original(*args, **kwargs)

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/workspace/ComfyUI/comfy/ldm/lightricks/model.py", line 975, in _forward

timestep, embedded_timestep, prompt_timestep = self._prepare_timestep(timestep, batch_size, input_dtype, **merged_args)

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/workspace/ComfyUI/custom_nodes/10S-Comfy-nodes/ltx_reference_enable.py", line 373, in _patched_prepare_timestep

return _ORIGINAL_PREPARE_TIMESTEP(self, timestep, batch_size, hidden_dtype, **kwargs)

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/workspace/ComfyUI/comfy/ldm/lightricks/av_model.py", line 775, in _prepare_timestep

av_ca_video_scale_shift_timestep, _ = self.av_ca_video_scale_shift_adaln_single(

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/venv/main/lib/python3.12/site-packages/torch/nn/modules/module.py", line 1779, in _wrapped_call_impl

return self._call_impl(*args, **kwargs)

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/venv/main/lib/python3.12/site-packages/torch/nn/modules/module.py", line 1790, in _call_impl

return forward_call(*args, **kwargs)

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/workspace/ComfyUI/comfy/ldm/lightricks/model.py", line 242, in forward

return self.linear(self.silu(embedded_timestep)), embedded_timestep

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/venv/main/lib/python3.12/site-packages/torch/nn/modules/module.py", line 1779, in _wrapped_call_impl

return self._call_impl(*args, **kwargs)

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/venv/main/lib/python3.12/site-packages/torch/nn/modules/module.py", line 1790, in _call_impl

return forward_call(*args, **kwargs)

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/workspace/ComfyUI/comfy/ops.py", line 1288, in forward

output = self.forward_comfy_cast_weights(

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/workspace/ComfyUI/comfy/ops.py", line 1230, in forward_comfy_cast_weights

x = self._forward(input, weight, bias)

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/workspace/ComfyUI/comfy/ops.py", line 1201, in _forward

return torch.nn.functional.linear(input, weight, bias)

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

torch.OutOfMemoryError: Allocation on device 0 would exceed allowed memory. (out of memory)

Currently allocated : 29.51 GiB

Requested : 550.00 MiB

Device limit : 31.36 GiB

Free (according to CUDA): 28.69 MiB

PyTorch limit (set by user-supplied memory fraction)

: 17179869184.00 GiB

[INFO] Memory summary:

|===========================================================================|

PyTorch CUDA memory summary, device ID 0
--
CUDA OOMs: 0
===========================================================================
Metric
---------------------------------------------------------------------------
Allocated memory
from large pool
from small pool
---------------------------------------------------------------------------
Active memory
from large pool
from small pool
---------------------------------------------------------------------------
Requested memory
from large pool
from small pool
---------------------------------------------------------------------------
GPU reserved memory
from large pool
from small pool
---------------------------------------------------------------------------
Non-releasable memory
from large pool
from small pool
---------------------------------------------------------------------------
Allocations
from large pool
from small pool
---------------------------------------------------------------------------
Active allocs
from large pool
from small pool
---------------------------------------------------------------------------
GPU reserved segments
from large pool
from small pool
---------------------------------------------------------------------------
Non-releasable allocs
from large pool
from small pool
---------------------------------------------------------------------------
Oversize allocations
---------------------------------------------------------------------------
Oversize GPU segments
===========================================================================

[ERROR] Got an OOM, unloading all loaded models.

[INFO] Prompt executed in 12.61 seconds

^C[INFO]

Stopped server

WORKFLOW: [Job ceb6b93e-32c1-49e8-8232-406eb2c1b9a6.json](https://github.com/user-attachments/files/29485603/Job.ceb6b93e-32c1-49e8-8232-406eb2c1b9a6.json)

from large pool | 0 MiB | 0 MiB | 0 B | 0 B |
| from small pool | 0 MiB | 0 MiB | 0 B | 0 B |
|---------------------------------------------------------------------------|
| Active memory | 29878 MiB | 30453 MiB | 0 B | 0 B |
| from large pool | 0 MiB | 0 MiB | 0 B | 0 B |
| from small pool | 0 MiB | 0 MiB | 0 B | 0 B |
|---------------------------------------------------------------------------|
| Requested memory | 0 B | 0 B | 0 B | 0 B |
| from large pool | 0 B | 0 B | 0 B | 0 B |
| from small pool | 0 B | 0 B | 0 B | 0 B |
|---------------------------------------------------------------------------|
| GPU reserved memory | 30880 MiB | 30880 MiB | 0 B | 0 B |
| from large pool | 0 MiB | 0 MiB | 0 B | 0 B |
| from small pool | 0 MiB | 0 MiB | 0 B | 0 B |
|---------------------------------------------------------------------------|
| Non-releasable memory | 0 B | 0 B | 0 B | 0 B |
| from large pool | 0 B | 0 B | 0 B | 0 B |
| from small pool | 0 B | 0 B | 0 B | 0 B |
|---------------------------------------------------------------------------|
| Allocations | 0 | 0 | 0 | 0 |
| from large pool | 0 | 0 | 0 | 0 |
| from small pool | 0 | 0 | 0 | 0 |
|---------------------------------------------------------------------------|
| Active allocs | 0 | 0 | 0 | 0 |
| from large pool | 0 | 0 | 0 | 0 |
| from small pool | 0 | 0 | 0 | 0 |
|---------------------------------------------------------------------------|
| GPU reserved segments | 0 | 0 | 0 | 0 |
| from large pool | 0 | 0 | 0 | 0 |
| from small pool | 0 | 0 | 0 | 0 |
|---------------------------------------------------------------------------|
| Non-releasable allocs | 0 | 0 | 0 | 0 |
| from large pool | 0 | 0 | 0 | 0 |
| from small pool | 0 | 0 | 0 | 0 |
|---------------------------------------------------------------------------|
| Oversize allocations | 0 | 0 | 0 | 0 |
|---------------------------------------------------------------------------|
| Oversize GPU segments | 0 | 0 | 0 | 0 |
|===========================================================================|

[ERROR] Got an OOM, unloading all loaded models.
[INFO] Prompt executed in 70.11 seconds
[INFO] Requested to load LTXAV
[INFO] loaded partially; 27620.41 MB usable, 27580.40 MB loaded, 1208.96 MB offloaded, 40.01 MB buffer reserved, lowvram patches: 407
[INFO] Patching torch settings: torch.backends.cuda.matmul.allow_fp16_accumulation = True
33%|████████████████████████████████████████████████████████▎ | 1/3 [00:04<00:09, 4.58s/it]
[INFO] Patching torch settings: torch.backends.cuda.matmul.allow_fp16_accumulation = False
→ [10S] TiledSampler v2.5: carrier wrapper sampling failed (OutOfMemoryError: Allocation on device 0 would exceed allowed memory. (out of memory)
Currently allocated : 29.60 GiB
Requested : 128.91 MiB
Device limit : 31.36 GiB
Free (according to CUDA): 28.69 MiB
PyTorch limit (set by user-supplied memory fraction)
: 17179869184.00 GiB); falling back to plain video for carrier tile
[INFO] 0 models unloaded.
[INFO] Unloaded partially: 635.17 MB freed, 26945.23 MB remains loaded, 40.01 MB buffer reserved, lowvram patches: 787
[INFO] Patching torch settings: torch.backends.cuda.matmul.allow_fp16_accumulation = True
0%| | 0/3 [00:00

Reproduce the attached Job ceb6b93e-32c1-49e8-8232-406eb2c1b9a6.json workflow with the reported frame counts and inspect the OOM path beginning in custom_nodes/10S-Comfy-nodes/latent_tiled_sampler.py. Trace the sampling calls through comfy/samplers.py into comfy/ldm/lightricks/av_model.py and model.py; done means the workflow completes without an OOM under the reported configuration.

Written by the indexing model from the issue text.