From e16473c6b53a37542dec4423a9fd82f7bdf7b92b Mon Sep 17 00:00:00 2001 From: Hongsheng Liu Date: Wed, 12 Aug 2026 22:24:58 +0800 Subject: [PATCH] docs: align VAE parallel file convention Signed-off-by: Hongsheng Liu --- docs/.nav.yml | 4 ++-- docs/design/architecture_overview.md | 2 +- docs/design/feature/{vae_parallelism.md => vae_parallel.md} | 0 docs/design/index.md | 2 +- docs/user_guide/diffusion/parallelism/overview.md | 2 +- .../parallelism/{vae_parallelism.md => vae_parallel.md} | 0 docs/user_guide/diffusion_features.md | 2 +- 7 files changed, 6 insertions(+), 6 deletions(-) rename docs/design/feature/{vae_parallelism.md => vae_parallel.md} (100%) rename docs/user_guide/diffusion/parallelism/{vae_parallelism.md => vae_parallel.md} (100%) diff --git a/docs/.nav.yml b/docs/.nav.yml index b1e3148b8..da3c10ecf 100644 --- a/docs/.nav.yml +++ b/docs/.nav.yml @@ -71,7 +71,7 @@ nav: - Hybrid Sharded Data Parallel: user_guide/diffusion/parallelism/hsdp.md - Sequence Parallel: user_guide/diffusion/parallelism/sequence_parallel.md - Tensor Parallel: user_guide/diffusion/parallelism/tensor_parallel.md - - VAE Parallelism: user_guide/diffusion/parallelism/vae_parallelism.md + - VAE Parallelism: user_guide/diffusion/parallelism/vae_parallel.md - Attention Backends: user_guide/diffusion/attention_backends.md - Regional Compilation: user_guide/diffusion/regional_compilation.md - Frame Interpolation: user_guide/diffusion/frame_interpolation.md @@ -123,7 +123,7 @@ nav: - Pipeline Parallel: design/feature/pipeline_parallel.md - Sequence Parallel: design/feature/sequence_parallel.md - Tensor Parallel: design/feature/tensor_parallel.md - - VAE Parallelism: design/feature/vae_parallelism.md + - VAE Parallelism: design/feature/vae_parallel.md - Attention Optimization: - Skip-Softmax: design/feature/skip_softmax.md - Cache-DiT: design/feature/cache_dit.md diff --git a/docs/design/architecture_overview.md b/docs/design/architecture_overview.md index ffc7292ec..a121069d9 100644 --- a/docs/design/architecture_overview.md +++ b/docs/design/architecture_overview.md @@ -398,7 +398,7 @@ the source for configuration, compatibility, and implementation details. * **Composable parallelism:** Diffusion stages can combine [CFG-Parallel](feature/cfg_parallel.md), [Expert Parallel](feature/expert_parallel.md), [HSDP](feature/hsdp.md), [Pipeline Parallel](feature/pipeline_parallel.md), [Sequence Parallel](feature/sequence_parallel.md), - [Tensor Parallel](feature/tensor_parallel.md), and [VAE Parallelism](feature/vae_parallelism.md) + [Tensor Parallel](feature/tensor_parallel.md), and [VAE Parallelism](feature/vae_parallel.md) according to the pipeline and hardware topology. * **Attention and cache acceleration:** [Skip-Softmax](feature/skip_softmax.md), [Cache-DiT](feature/cache_dit.md), and [TeaCache](feature/teacache.md) diff --git a/docs/design/feature/vae_parallelism.md b/docs/design/feature/vae_parallel.md similarity index 100% rename from docs/design/feature/vae_parallelism.md rename to docs/design/feature/vae_parallel.md diff --git a/docs/design/index.md b/docs/design/index.md index b03eb0ba7..eca3ecd3b 100644 --- a/docs/design/index.md +++ b/docs/design/index.md @@ -48,7 +48,7 @@ implementation contract; it is not, by itself, a general support claim. - [Pipeline Parallel](feature/pipeline_parallel.md) - [Sequence Parallel](feature/sequence_parallel.md) - [Tensor Parallel](feature/tensor_parallel.md) -- [VAE Parallelism](feature/vae_parallelism.md) +- [VAE Parallelism](feature/vae_parallel.md) #### Attention optimization diff --git a/docs/user_guide/diffusion/parallelism/overview.md b/docs/user_guide/diffusion/parallelism/overview.md index 8419b4800..362492f45 100644 --- a/docs/user_guide/diffusion/parallelism/overview.md +++ b/docs/user_guide/diffusion/parallelism/overview.md @@ -10,7 +10,7 @@ This guide covers the parallelism methods in vLLM-Omni for speeding up diffusion | **[Sequence Parallelism](sequence_parallel.md)** | Splits sequence dimension across GPUs (Ulysses-SP, Ring-Attention, or hybrid) for high-resolution images and videos | | **[CFG-Parallel](cfg_parallel.md)** | Runs CFG positive/negative branches on separate GPUs for ~1.8x speedup on guided generation | | **[Pipeline Parallelism](pipeline_parallel.md)** | Splits the denoising transformer block-wise across sequential GPU stages to reduce per-GPU model memory | -| **[VAE Parallelism](vae_parallelism.md)** | Distributes VAE decode spatially across GPUs to reduce peak VAE memory | +| **[VAE Parallelism](vae_parallel.md)** | Distributes VAE decode spatially across GPUs to reduce peak VAE memory | | **[HSDP](hsdp.md)** | Shards full model weights via PyTorch FSDP2 to enable large-model inference on memory-constrained GPUs | | **[Expert Parallelism](expert_parallel.md)** | Shards MoE expert blocks across GPUs for MoE models (e.g. HunyuanImage3.0) | diff --git a/docs/user_guide/diffusion/parallelism/vae_parallelism.md b/docs/user_guide/diffusion/parallelism/vae_parallel.md similarity index 100% rename from docs/user_guide/diffusion/parallelism/vae_parallelism.md rename to docs/user_guide/diffusion/parallelism/vae_parallel.md diff --git a/docs/user_guide/diffusion_features.md b/docs/user_guide/diffusion_features.md index 7c664e88b..e2b42916e 100644 --- a/docs/user_guide/diffusion_features.md +++ b/docs/user_guide/diffusion_features.md @@ -61,7 +61,7 @@ Memory optimization methods help reduce GPU memory usage, enabling inference on |--------|-------------|----------| | **[CPU Offload](diffusion/cpu_offload.md)** | Offloads model components to CPU memory | Limited VRAM, large models on consumer GPUs | | **[Quantization](quantization/overview.md)** | Reduces transformer stages from BF16 to FP8/INT8/etc. | Limited VRAM, minimal accuracy loss | -| **[VAE Parallelism](diffusion/parallelism/vae_parallelism.md)** | Distributes VAE decode work across GPUs | High-resolution generation with reduced VAE memory peak | +| **[VAE Parallelism](diffusion/parallelism/vae_parallel.md)** | Distributes VAE decode work across GPUs | High-resolution generation with reduced VAE memory peak | ### Extensions