Skip to content

[Feat] support HSDP for Flux family #1900

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
gcanlin merged 3 commits into from
Mar 15, 2026
Merged

[Feat] support HSDP for Flux family #1900

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
8ffbf84
[Feat] support HSDP for Flux family
RuixiangMa Mar 14, 2026
110d846
Merge branch 'main' into fluxhsdp
RuixiangMa Mar 15, 2026
aa23e7f
upd
RuixiangMa Mar 15, 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
32 changes: 16 additions & 16 deletions docs/user_guide/diffusion/parallelism_acceleration.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -24,22 +24,22 @@ The following table shows which models are currently supported by parallelism me

### ImageGen

| Model | Model Identifier | Ulysses-SP | Ring-SP | CFG-Parallel | Tensor-Parallel | VAE-Patch-Parallel | Expert-Parallel |
|--------------------------|--------------------------------------|:----------:|:-------:|:------------:|:---------------:|:------------------:|:---------------:|
| **LongCat-Image** | `meituan-longcat/LongCat-Image` | ✅ | ✅ | ❌ | ✅ | ❌ | N/A |
| **LongCat-Image-Edit** | `meituan-longcat/LongCat-Image-Edit` | ✅ | ✅ | ❌ | ✅ | ❌ | N/A |
| **Ovis-Image** | `OvisAI/Ovis-Image` | ❌ | ❌ | ❌ | ❌ | ❌ | N/A |
| **Qwen-Image** | `Qwen/Qwen-Image` | ✅ | ✅ | ✅ | ✅ | ✅ | N/A |
| **Qwen-Image-Edit** | `Qwen/Qwen-Image-Edit` | ✅ | ✅ | ✅ | ✅ | ❌ | N/A |
| **Qwen-Image-Edit-2509** | `Qwen/Qwen-Image-Edit-2509` | ✅ | ✅ | ✅ | ✅ | ❌ | N/A |
| **Qwen-Image-Layered** | `Qwen/Qwen-Image-Layered` | ✅ | ✅ | ✅ | ✅ | ❌ | N/A |
| **Z-Image** | `Tongyi-MAI/Z-Image-Turbo` | ✅ | ✅ | ❌ | ✅ (TP=2 only) | ✅ | N/A |
| **Stable-Diffusion3.5** | `stabilityai/stable-diffusion-3.5` | ❌ | ❌ | ❌ | ✅ | ✅ | N/A |
| **FLUX.2-klein** | `black-forest-labs/FLUX.2-klein-4B` | ✅ | ✅ | ❌ | ✅ | ❌ | N/A |
| **FLUX.1-dev** | `black-forest-labs/FLUX.1-dev` | ❌ | ❌ | ✅ | ✅ | ❌ | N/A |
| **FLUX.2-dev** | `black-forest-labs/FLUX.2-dev` | ❌ | ❌ | ❌ | ✅ | ❌ | N/A |
| **HunyuanImage3.0** | `tencent/HunyuanImage-3.0`, `tencent/HunyuanImage-3.0-Instruct` | ❌ | ❌ | ❌ | ✅ | ❌ | ✅ |
| **DreamID-Omni** | `XuGuo699/DreamID-Omni` | ❌ | ❌ | ✅ | ❌ | ❌ | N/A |
| Model | Model Identifier | Ulysses-SP | Ring-SP | CFG-Parallel | Tensor-Parallel | VAE-Patch-Parallel | Expert-Parallel | HSDP |
|--------------------------|--------------------------------------|:----------:|:-------:|:------------:|:---------------:|:------------------:|:---------------:|:----:|
| **LongCat-Image** | `meituan-longcat/LongCat-Image` | ✅ | ✅ | ❌ | ✅ | ❌ | N/A | ❌ |
| **LongCat-Image-Edit** | `meituan-longcat/LongCat-Image-Edit` | ✅ | ✅ | ❌ | ✅ | ❌ | N/A | ❌ |
| **Ovis-Image** | `OvisAI/Ovis-Image` | ❌ | ❌ | ❌ | ❌ | ❌ | N/A | ❌ |
| **Qwen-Image** | `Qwen/Qwen-Image` | ✅ | ✅ | ✅ | ✅ | ✅ | N/A | ❌ |
| **Qwen-Image-Edit** | `Qwen/Qwen-Image-Edit` | ✅ | ✅ | ✅ | ✅ | ❌ | N/A | ❌ |
| **Qwen-Image-Edit-2509** | `Qwen/Qwen-Image-Edit-2509` | ✅ | ✅ | ✅ | ✅ | ❌ | N/A | ❌ |
| **Qwen-Image-Layered** | `Qwen/Qwen-Image-Layered` | ✅ | ✅ | ✅ | ✅ | ❌ | N/A | ❌ |
| **Z-Image** | `Tongyi-MAI/Z-Image-Turbo` | ✅ | ✅ | ❌ | ✅ (TP=2 only) | ✅ | N/A | ❌ |
| **Stable-Diffusion3.5** | `stabilityai/stable-diffusion-3.5` | ❌ | ❌ | ❌ | ✅ | ✅ | N/A | ❌ |
| **FLUX.2-klein** | `black-forest-labs/FLUX.2-klein-4B` | ✅ | ✅ | ❌ | ✅ | ❌ | N/A | ✅ |
| **FLUX.1-dev** | `black-forest-labs/FLUX.1-dev` | ❌ | ❌ | ✅ | ✅ | ❌ | N/A | ✅ |
| **FLUX.2-dev** | `black-forest-labs/FLUX.2-dev` | ❌ | ❌ | ❌ | ✅ | ❌ | N/A | ✅ |
| **HunyuanImage3.0** | `tencent/HunyuanImage-3.0`, `tencent/HunyuanImage-3.0-Instruct` | ❌ | ❌ | ❌ | ✅ | ❌ | ✅ | ❌ |
| **DreamID-Omni** | `XuGuo699/DreamID-Omni` | ❌ | ❌ | ✅ | ❌ | ❌ | N/A | ❌ |

!!! note "TP Limitations for Diffusion Models"
We currently implement Tensor Parallelism (TP) only for the DiT (Diffusion Transformer) blocks. This is because the `text_encoder` component in vLLM-Omni uses the original Transformers implementation, which does not yet support TP.
Expand Down
6 changes: 6 additions & 0 deletions vllm_omni/diffusion/models/flux/flux_transformer.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -470,6 +470,12 @@ class FluxTransformer2DModel(nn.Module):
# -- typically a transformer layer
# used for torch compile optimizations
_repeated_blocks = ["FluxTransformerBlock"]

@staticmethod
def _is_transformer_block(name: str, module) -> bool:
return ("transformer_blocks" in name or "single_transformer_blocks" in name) and name.split(".")[-1].isdigit()

_hsdp_shard_conditions = [_is_transformer_block]
packed_modules_mapping = {
"to_qkv": ["to_q", "to_k", "to_v"],
"add_kv_proj": ["add_q_proj", "add_k_proj", "add_v_proj"],
Expand Down
4 changes: 2 additions & 2 deletions vllm_omni/diffusion/models/flux/pipeline_flux.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -160,10 +160,10 @@ def __init__(
)
self.text_encoder = CLIPTextModel.from_pretrained(
model, subfolder="text_encoder", local_files_only=local_files_only
)
).to(self.device)
self.text_encoder_2 = T5EncoderModel.from_pretrained(
model, subfolder="text_encoder_2", local_files_only=local_files_only
)
).to(self.device)
self.vae = AutoencoderKL.from_pretrained(model, subfolder="vae", local_files_only=local_files_only).to(
self.device
)
Expand Down
6 changes: 6 additions & 0 deletions vllm_omni/diffusion/models/flux2/flux2_transformer.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -553,6 +553,12 @@ class Flux2Transformer2DModel(nn.Module):

_repeated_blocks = ["Flux2TransformerBlock", "Flux2SingleTransformerBlock"]

@staticmethod
def _is_transformer_block(name: str, module) -> bool:
return ("transformer_blocks" in name or "single_transformer_blocks" in name) and name.split(".")[-1].isdigit()

_hsdp_shard_conditions = [_is_transformer_block]

def __init__(
self,
patch_size: int = 1,
Expand Down
2 changes: 1 addition & 1 deletion vllm_omni/diffusion/models/flux2/pipeline_flux2.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -366,7 +366,7 @@ def __init__(
)
self.text_encoder = Mistral3ForConditionalGeneration.from_pretrained(
model, subfolder="text_encoder", local_files_only=local_files_only
)
).to(self._execution_device)
self.tokenizer = PixtralProcessor.from_pretrained(
model, subfolder="tokenizer", local_files_only=local_files_only
)
Expand Down
5 changes: 5 additions & 0 deletions vllm_omni/diffusion/models/flux2_klein/flux2_klein_transformer.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -741,6 +741,11 @@ class Flux2Transformer2DModel(nn.Module):

_repeated_blocks = ["Flux2TransformerBlock", "Flux2SingleTransformerBlock"]

@staticmethod
def _is_transformer_block(name: str, module) -> bool:
return ("transformer_blocks" in name or "single_transformer_blocks" in name) and name.split(".")[-1].isdigit()

_hsdp_shard_conditions = [_is_transformer_block]
_sp_plan = {
"": {
"hidden_states": SequenceParallelInput(split_dim=1, expected_dims=3, auto_pad=True),
Expand Down
2 changes: 1 addition & 1 deletion vllm_omni/diffusion/models/flux2_klein/pipeline_flux2_klein.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -218,7 +218,7 @@ def __init__(
model,
subfolder="text_encoder",
local_files_only=local_files_only,
)
).to(self._execution_device)
self.tokenizer = Qwen2TokenizerFast.from_pretrained(
model,
subfolder="tokenizer",
Expand Down
Loading