diff --git a/docs_new/docs.json b/docs_new/docs.json index 1372ec599..a877b7135 100644 --- a/docs_new/docs.json +++ b/docs_new/docs.json @@ -1158,7 +1158,6 @@ "group": "References", "pages": [ "docs/sglang-diffusion/environment_variables", - "docs/sglang-diffusion/compatibility_matrix", "docs/sglang-diffusion/support_new_models", "docs/sglang-diffusion/contributing" ] diff --git a/docs_new/docs/sglang-diffusion/compatibility_matrix.mdx b/docs_new/docs/sglang-diffusion/compatibility_matrix.mdx index 43935a98e..b5bf1d6be 100644 --- a/docs_new/docs/sglang-diffusion/compatibility_matrix.mdx +++ b/docs_new/docs/sglang-diffusion/compatibility_matrix.mdx @@ -18,31 +18,41 @@ default parameters when initializing and generating videos. ### Video Generation Models +Optimization columns are abbreviated to keep the matrix readable: + +- `Tea` = TeaCache +- `Tile` = Sliding Tile Attention +- `Sage` = Sage Attention +- `VSA` = Video Sparse Attention +- `SLA` = Sparse Linear Attention +- `SageSLA` = Sage Sparse Linear Attention +- `SVG2` = Sparse Video Gen 2 + - + - - - - - - - + + + + + + + - - - - - - - - + + + + + + + + @@ -85,7 +95,7 @@ default parameters when initializing and generating videos. - + @@ -97,7 +107,7 @@ default parameters when initializing and generating videos. - + @@ -109,7 +119,7 @@ default parameters when initializing and generating videos. - + @@ -121,7 +131,7 @@ default parameters when initializing and generating videos. - + @@ -277,7 +287,7 @@ default parameters when initializing and generating videos. - + @@ -289,7 +299,7 @@ default parameters when initializing and generating videos. - + @@ -306,7 +316,7 @@ default parameters when initializing and generating videos. 1. Wan2.2 TI2V 5B has some quality issues when performing I2V generation. We are working on fixing this issue. 2. SageSLA is based on SpargeAttn. Install it first with `pip install git+https://github.com/thu-ml/SpargeAttn.git --no-build-isolation` 3. LTX-2 and LTX-2.3 two-stage generation uses `--pipeline-class-name LTX2TwoStagePipeline`. The spatial upsampler and distilled LoRA are auto-resolved from the model snapshot by default, and can still be overridden with `--spatial-upsampler-path` and `--distilled-lora-path`. - - For LTX models, the `Resolutions` column uses output video `width×height` semantics, matching `sglang generate --width ... --height ...`. + - For LTX models, the `Resolution` column uses output video `width×height` semantics, matching `sglang generate --width ... --height ...`. 4. LTX-2.3 two-stage also supports `--ltx2-two-stage-device-mode {legacy,snapshot,resident}`: - `snapshot` is the default and recommended mode. - `resident` usually provides the best latency/throughput but uses much more VRAM. diff --git a/docs_new/docs/sglang-diffusion/index.mdx b/docs_new/docs/sglang-diffusion/index.mdx index 570a273d2..d473d1891 100644 --- a/docs_new/docs/sglang-diffusion/index.mdx +++ b/docs_new/docs/sglang-diffusion/index.mdx @@ -29,22 +29,22 @@ sglang serve --model-path Qwen/Qwen-Image --port 30010 ## Start Here -- [Installation](./installation): install SGLang Diffusion and platform dependencies -- [Compatibility Matrix](./compatibility_matrix): check model, optimization, and component override support -- [CLI](./api/cli): run one-off generation jobs or launch a persistent server -- [OpenAI-Compatible API](./api/openai_api): send image and video requests to the HTTP server -- [Attention Backends](./attention_backends): choose the best backend for your model and hardware -- [Caching Acceleration](./caching-acceleration): use Cache-DiT or TeaCache to reduce denoising cost -- [Quantization](./quantization): load quantized transformer checkpoints -- [Contributing](./contributing): contribution workflow, adding new models, and CI perf baselines +- [Installation](/docs/sglang-diffusion/installation): install SGLang Diffusion and platform dependencies +- [Compatibility Matrix](/docs/sglang-diffusion/compatibility_matrix): check model, optimization, and component override support +- [CLI](/docs/sglang-diffusion/api/cli): run one-off generation jobs or launch a persistent server +- [OpenAI-Compatible API](/docs/sglang-diffusion/api/openai_api): send image and video requests to the HTTP server +- [Attention Backends](/docs/sglang-diffusion/attention_backends): choose the best backend for your model and hardware +- [Caching Acceleration](/docs/sglang-diffusion/caching-acceleration): use Cache-DiT or TeaCache to reduce denoising cost +- [Quantization](/docs/sglang-diffusion/quantization): load quantized transformer checkpoints +- [Contributing](/docs/sglang-diffusion/contributing): contribution workflow, adding new models, and CI perf baselines ## Additional Documentation -- [Post-Processing](./api/post_processing): frame interpolation and upscaling -- [Performance Overview](./performance-optimization): overview of attention, caching, and profiling -- [Environment Variables](./environment_variables): platform, caching, storage, and debugging configuration -- [Support New Models](./support_new_models): implementation guide for new diffusion pipelines -- [CI Performance](./ci_perf): performance baseline generation +- [Post-Processing](/docs/sglang-diffusion/api/post_processing): frame interpolation and upscaling +- [Performance Overview](/docs/sglang-diffusion/performance-optimization): overview of attention, caching, and profiling +- [Environment Variables](/docs/sglang-diffusion/environment_variables): platform, caching, storage, and debugging configuration +- [Support New Models](/docs/sglang-diffusion/support_new_models): implementation guide for new diffusion pipelines +- [CI Performance](/docs/sglang-diffusion/ci_perf): performance baseline generation ## References
Model Name Hugging Face Model IDResolutionsTeaCacheSliding Tile AttnSage AttnVideo Sparse Attention (VSA)Sparse Linear Attention (SLA)Sage Sparse Linear Attention (SageSLA)Sparse Video Gen 2 (SVG2)ResolutionTeaTileSageVSASLASageSLASVG2
Wan2.2 T2V A14B `Wan-AI/Wan2.2-T2V-A14B-Diffusers`480p<br>720p480p
720p		 ❌ ❌ ✅
Wan2.2 I2V A14B `Wan-AI/Wan2.2-I2V-A14B-Diffusers`480p<br>720p480p
720p		 ❌ ❌ ✅
HunyuanVideo `hunyuanvideo-community/HunyuanVideo`720×1280<br>544×960720×1280
544×960		 ❌ ✅ ✅
FastHunyuan `FastVideo/FastHunyuan-diffusers`720×1280<br>544×960720×1280
544×960		 ❌ ✅ ✅
LTX-2 (one and two stages) Lightricks/LTX-2768×512<br>1536×1024768×512
1536×1024		 ❌ ❌ ❌
LTX-2.3 (one and two stages) Lightricks/LTX-2.3768×512<br>1536×1024768×512
1536×1024		 ❌ ❌ ❌