Skip to content

Optimize Helios docs #13222

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
yiyixuxu merged 2 commits into from
Mar 8, 2026
+3 −4
Merged

Optimize Helios docs #13222

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
fe7bc4a
optimize helios docs
SHYuanBest Mar 8, 2026
151dd5a
Merge branch 'main' into fix-helios-docs
SHYuanBest Mar 8, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
7 changes: 3 additions & 4 deletions docs/source/en/api/pipelines/helios.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -44,7 +44,7 @@ The example below demonstrates how to generate a video from text optimized for m

Refer to the [Reduce memory usage](../../optimization/memory) guide for more details about the various memory saving techniques.

The Helios model below requires ~19GB of VRAM.
The Helios model below requires ~6GB of VRAM.

```py
import torch
Expand All @@ -63,8 +63,7 @@ pipeline = HeliosPipeline.from_pretrained(
pipeline.enable_group_offload(
onload_device=torch.device("cuda"),
offload_device=torch.device("cpu"),
offload_type="block_level",
num_blocks_per_group=1,
offload_type="leaf_level",
use_stream=True,
record_stream=True,
)
Expand Down Expand Up @@ -97,7 +96,7 @@ export_to_video(output, "helios_base_t2v_output.mp4", fps=24)
</hfoption>
<hfoption id="inference speed">

[Compilation](../../optimization/fp16#torchcompile) is slow the first time but subsequent calls to the pipeline are faster. [Attention Backends](../../optimization/attention_backends) such as FlashAttention and SageAttention can significantly increase speed by optimizing the computation of the attention mechanism. [Caching](../../optimization/cache) may also speed up inference by storing and reusing intermediate outputs.
[Compilation](../../optimization/fp16#torchcompile) is slow the first time but subsequent calls to the pipeline are faster. [Attention Backends](../../optimization/attention_backends) such as FlashAttention and SageAttention can significantly increase speed by optimizing the computation of the attention mechanism. [Context Parallelism](../../training/distributed_inference#context-parallelism) splits the input sequence across multiple devices to enable processing of long contexts in parallel, reducing memory pressure and latency. [Caching](../../optimization/cache) may also speed up inference by storing and reusing intermediate outputs.

```py
import torch
Expand Down