Skip to content
Pruna AI Customer Support Portal home
Pruna AI Customer Support Portal home

BRIA3.2

By the time BRIA 3.2 launched publicly at NVIDIA GTC Paris x VivaTech, Pruna delivered multiple runtime configurations offering 2x to 3.6x faster inference, with quality evaluation validated by BRIA’s in-house visual team.

All tests were run on L40S GPUs (i.e., AWS g6e.2xlarge instances), the same hardware family that BRIA uses in production, so results were apples to apples. For quality, we used LPIPS (captures perceptual similarity as humans see it), SSIM (compares image structure and texture), and PSNR (detects pixel-level distortion) — three metrics that together gave a reliable picture of visual fidelity between the base and optimized model.

Try it on your setup.

import torch from huggingface_hub import hf_hub_download from pruna_pro import SmashConfig, smash # download BRIA-3.2 repo_id = "briaai/BRIA-3.2" hf_hub_download(repo_id=repo_id, filename="pipeline_bria.py") hf_hub_download(repo_id=repo_id, filename="transformer_bria.py") hf_hub_download(repo_id=repo_id, filename="bria_utils.py") from pipeline_bria import BriaPipeline pipe = BriaPipeline.from_pretrained( "briaai/BRIA-3.2", torch_dtype=torch.bfloat16, trust_remote_code=True, ).to(device="cuda") # smash BRIA-3.2 with Pruna Pro smash_config = SmashConfig() smash_config["compiler"] = "torch_compile" smash_config["torch_compile_target"] = "module_list" smash_config["cacher"] = "auto" smash_config["auto_cache_mode"] = "taylor" smash_config["auto_speed_factor"] = 0.7 # 0.5 for even faster inference smash_config._prepare_saving = False pipe = smash(pipe, smash_config, experimental=True) # run inference with smashed pipe prompt = "A portrait of a Beautiful and playful ethereal singer, golden designs, highly detailed, blurry background" negative_prompt = "Logo,Watermark,Ugly,Morbid,Extra fingers,Poorly drawn hands,Mutation,Blurry,Extra limbs,Gross proportions,Missing arms,Mutated hands,Long neck,Duplicate,Mutilated,Mutilated hands,Poorly drawn face,Deformed,Bad anatomy,Cloned face,Malformed limbs,Missing legs,Too many fingers" images = pipe(prompt=prompt, negative_prompt=negative_prompt, height=1024, width=1024).images[0]

Read the full benchmark: https://www.pruna.ai/blog/4x-faster-bria32