From fe7e8770db43a845c4ea0407ddc61f17ba3a7f63 Mon Sep 17 00:00:00 2001 From: JohnJan Date: Thu, 12 Mar 2026 19:53:47 +0800 Subject: [PATCH] [Feature]: support Flux.2-dev cache_dit (#1814) Co-authored-by: wuzhongjian Co-authored-by: Gao Han --- docs/user_guide/diffusion_acceleration.md | 2 +- .../diffusion/cache/cache_dit_backend.py | 75 +++++++++++++++++++ 2 files changed, 76 insertions(+), 1 deletion(-) diff --git a/docs/user_guide/diffusion_acceleration.md b/docs/user_guide/diffusion_acceleration.md index ef4b39d0a..7f32e3eff 100644 --- a/docs/user_guide/diffusion_acceleration.md +++ b/docs/user_guide/diffusion_acceleration.md @@ -69,7 +69,7 @@ The following table shows which models are currently supported by each accelerat | **FLUX.1-dev** | `black-forest-labs/FLUX.1-dev` | ❌ | ✅ | ❌ | ❌ | ✅ | ✅ | ❌ | | **NextStep-1.1** | `stepfun-ai/NextStep-1.1` | ❌ | ❌ | ❌ | ❌ | ✅ | ✅ | ❌ | | **FLUX.2-klein** | `black-forest-labs/FLUX.2-klein-4B` | ❌ | ✅ | ❌ | ❌ | ❌ | ✅ | ❌ | -| **FLUX.2-dev** | `black-forest-labs/FLUX.2-dev` | ❌ | ❌ | ❌ | ❌ | ❌ | ✅ | ❌ | +| **FLUX.2-dev** | `black-forest-labs/FLUX.2-dev` | ❌ | ✅ | ❌ | ❌ | ❌ | ✅ | ❌ | ### VideoGen diff --git a/vllm_omni/diffusion/cache/cache_dit_backend.py b/vllm_omni/diffusion/cache/cache_dit_backend.py index 1c98d0724..d7165eda2 100644 --- a/vllm_omni/diffusion/cache/cache_dit_backend.py +++ b/vllm_omni/diffusion/cache/cache_dit_backend.py @@ -984,6 +984,80 @@ def enable_cache_for_bagel(pipeline: Any, cache_config: Any) -> Callable[[int], return refresh_cache_context +def enable_cache_for_flux2(pipeline: Any, cache_config: Any) -> Callable[[int], None]: + """Enable cache-dit for Flux.2-dev pipeline. + + Args: + pipeline: The Flux2 pipeline instance. + cache_config: DiffusionCacheConfig instance with cache configuration. + Returns: + A refresh function that can be called with a new ``num_inference_steps`` + to update the cache context for the pipeline. + """ + # Build DBCacheConfig for transformer + db_cache_config = _build_db_cache_config(cache_config) + + calibrator = None + if cache_config.enable_taylorseer: + taylorseer_order = cache_config.taylorseer_order + calibrator = TaylorSeerCalibratorConfig(taylorseer_order=taylorseer_order) + logger.info(f"TaylorSeer enabled with order={taylorseer_order}") + + # Build ParamsModifier for transformer + modifier = ParamsModifier( + cache_config=db_cache_config, + calibrator_config=calibrator, + ) + + logger.info( + f"Enabling cache-dit on Flux transformer with BlockAdapter: " + f"Fn={db_cache_config.Fn_compute_blocks}, " + f"Bn={db_cache_config.Bn_compute_blocks}, " + f"W={db_cache_config.max_warmup_steps}, " + ) + + # Enable cache-dit using BlockAdapter for transformer + cache_dit.enable_cache( + ( + BlockAdapter( + transformer=pipeline.transformer, + blocks=[ + pipeline.transformer.transformer_blocks, + pipeline.transformer.single_transformer_blocks, + ], + forward_pattern=[ForwardPattern.Pattern_1, ForwardPattern.Pattern_2], + params_modifiers=[modifier], + ) + ), + cache_config=db_cache_config, + ) + + def refresh_cache_context(pipeline: Any, num_inference_steps: int, verbose: bool = True) -> None: + """Refresh cache context for the transformer with new num_inference_steps. + + Args: + pipeline: The Flux2 pipeline instance. + num_inference_steps: New number of inference steps. + """ + if cache_config.scm_steps_mask_policy is None: + cache_dit.refresh_context(pipeline.transformer, num_inference_steps=num_inference_steps, verbose=verbose) + else: + cache_dit.refresh_context( + pipeline.transformer, + cache_config=DBCacheConfig().reset( + num_inference_steps=num_inference_steps, + steps_computation_mask=cache_dit.steps_mask( + mask_policy=cache_config.scm_steps_mask_policy, + total_steps=num_inference_steps, + ), + steps_computation_policy=cache_config.scm_steps_policy, + ), + verbose=verbose, + ) + + return refresh_cache_context + + # Register custom cache-dit enablers after function definitions CUSTOM_DIT_ENABLERS.update( { @@ -998,6 +1072,7 @@ CUSTOM_DIT_ENABLERS.update( "LTX2Pipeline": enable_cache_for_ltx2, "LTX2ImageToVideoPipeline": enable_cache_for_ltx2, "BagelPipeline": enable_cache_for_bagel, + "Flux2Pipeline": enable_cache_for_flux2, } )