You are viewing an old version of this page. View the current version.

Compare with Current View Page History

« Previous Version 11 Next »

Info


DRAFT ARTICLE


https://huggingface.co/tencent/SRPO

Test 0 - Different seed variations

Prompt: photorealistic girl in bookshop choosing the book in romantic stories shelf. smiling

Parameters: Steps: 50| Size: 1024x1024| Seed: 1620085323| CFG scale: 6| App: SD.Next| Version: 82504e9| Pipeline: FluxPipeline| Operations: txt2img| Model: flux.1-dev-SRPO

Time: 27m 11.72s | total 1246.29 pipeline 1224.28 preview 7.57 decode 6.36 callback 3.69 prompt 2.02 te 2.01 gc 0.33 | GPU 34676 MB 29% | RAM 4.04 GB 3%


CFG6, STEP 32Seed: 1620085323Seed:1931701040Seed:4075624134Seed:2736029172
bookshop girl

hand and face

legs and shoes

Test 1 - Bookshop

Prompt: photorealistic girl in bookshop choosing the book in romantic stories shelf. smiling


48162032

CFG1

CFG2


CFG3






CFG4






CFG5






CFG6






CFG8







Test 2 - Face and hand



8162032

CFG1





CFG2





CFG3





CFG3.5





CFG4





CFG5





CFG8





Test 3 - Legs

Prompt: Generate a photo of a woman's legs, with her feet crossed and wearing white high-heeled shoes with ribbons tied around her ankles. The shoes should have a pointed toe and a stiletto heel. The woman's legs should be smooth and tanned, with a slight sheen to them. The background should be a light gray color. The photo should be taken from a low angle, looking up at the woman's legs. The ribbons should be tied in a bow shape around the ankles. The shoes should have a red sole. The woman's legs should be slightly bent at the knee.



8162032

CFG1





CFG2





CFG3





CFG3.5





CFG4





CFG5





CFG8





Test 4 CivitAi profile cover generation

1600x400


Prompt: image with 5 canvases and one android robot painting on the 3rd canvas. first two is done with some futuristic images, last two is blank. robot is surprised look back at the camera. paint brush is in the robot hand. Left bottom corner text "Flux.1 Krea"

Parameters: Steps: 32| Size: 1600x400| Seed: 433989387| CFG scale: 3.5


Prompt: image with 5 canvases and one android robot painting on the 3rd canvas. first two is done with some futuristic images, last two is blank. robot is surprised look back at the camera. paint brush is in the robot hand. Left bottom corner text "Flux.1 dev"

Parameters: Steps: 32| Size: 1600x400| Seed: 4014230639| CFG scale: 8


Prompt: with analog noise and glitches, CivitAi cover image with 5 canvases and one android robot painting on the 3rd canvas. first two is done with some futuristic images, last two is blank. robot is surprised look back at the camera. paint brush is in the robot hand.

Parameters: Steps: 32| Size: 1600x400| Seed: 2404111820| CFG scale: 3.5


Prompt: with analog noise and glitches, CivitAi cover image with 5 canvases and one android robot painting on the 3rd canvas. first two is done with some futuristic images, last two is blank. robot is surprised look back at the camera. paint brush is in the robot hand. Left bottom corner text "SD.Next + Flux.1 Krea dev", upper right corner text "Stable Diffusion at home"

Parameters: Steps: 32| Size: 1600x400| Seed: 109351024| CFG scale: 3.5


Prompt: 80s cartoon style, CivitAi cover image with 5 canvases and one android robot painting on the 3rd canvas. first two is done with some futuristic images, last two is blank. robot is suprised look back at the camera. paint brush is in the robot hand. Left bottom corner text "SD.Next + Flux.1 dev", uper right corner text "Stable Diffusion at home"

Parameters: Steps: 20| Size: 1600x400| Seed: 1620085323| CFG scale: 6


Prompt: 3d render style, CivitAi cover image with 5 canvases and one android robot painting on the 3rd canvas. first two is done with some futuristic images, last two is blank. robot is suprised look back at the camera. paint brush is in the robot hand. Left bottom corner text "SD.Next + Flux.1 dev", uper right corner text "Stable Diffusion at home"

Parameters: Steps: 20| Size: 1600x400| Seed: 1620085323| CFG scale: 6


Prompt: Retro poster style, CivitAi cover image with 5 canvases and one android robot painting on the 3rd canvas. first two is done with some futuristic images, last two is blank. robot is suprised look back at the camera. paint brush is in the robot hand. Left bottom corner text "SD.Next + Flux.1 dev", uper right corner text "Stable Diffusion at home"

Parameters: Steps: 20| Size: 1600x400| Seed: 1620085323| CFG scale: 6


Prompt: LEGO style, CivitAi cover image with 5 canvases and one android robot painting on the 3rd canvas. first two is done with some futuristic images, last two is blank. robot is surprised look back at the camera. paint brush is in the robot hand. Left bottom corner text "SD.Next + Flux.1 dev", upper right corner text "Stable Diffusion at home"

Parameters: Steps: 20| Size: 1600x400| Seed: 1620085323| CFG scale: 6


Prompt: CivitAi cover image with 5 canvases and one android robot painting on the 3rd canvas. first two is done with some futuristic images, last two is blank. robot is surprised look back at the camera. paint brush is in the robot hand. Left bottom corner text "SD.Next + Flux.1 dev", upper right corner text "Stable Diffusion at home"

Parameters: Steps: 20| Size: 1600x400| Seed: 1620085323| CFG scale: 6

Test 5 CFG 6 vs CFG 3.5

https://imgsli.com/NDE3MjY5

https://imgsli.com/NDE3Mjcx

CFG3.5

CFG 6

System info


Mon Sep 22 09:38:44 2025
app: sdnext.git updated: 2025-09-21 hash: 82504e96 url: https://github.com/vladmandic/sdnext.git/tree/dev
arch: x86_64 cpu: x86_64 system: Linux release: 6.14.0-29-generic python: 3.12.3

Torch: 2.7.1+xpu device: Intel(R) Arc(TM) Graphics (1) ipex: 2.7.10+xpu

ram: free:121.28 used:4.05 total:125.33
gpu: free:81.62 used:35.75 total:117.37
gpu-active: current:31.63 peak:33.68
gpu-allocated: current:31.63 peak:33.68
gpu-reserved: current:35.75 peak:35.75
gpu-inactive: current:0.13 peak:0.74
events: retries:0 oom:0
utilization: 0

xformers: diffusers: 0.36.0.dev0 transformers: 4.56.1
active: xpu dtype: torch.bfloat16 vae: torch.bfloat16 unet: torch.bfloat16
base: vladmandic/flux.1-dev-SRPO refiner: none vae: none te: none unet: none
Backend: ipex Cross-attention: Scaled-Dot-Product


Config

{
  "theme_type": "Standard",
  "diffusers_version": "efb7a299af46d739dec6a57a5d2814165fba24b5",
  "diffusers_offload_mode": "none",
  "ui_request_timeout": 300000,
  "huggingface_token": "hf_..FraU",
  "samples_filename_pattern": "[date]-[seq]-[model_name]-[height]x[width]-Seed[seed]-CFG[cfg]-AG[pag]-STEP[steps]",
  "sd_model_checkpoint": "vladmandic/flux.1-dev-SRPO",
  "sd_checkpoint_hash": null,
  "hf_transfer_mode": "xet",
  "extra_networks_sort": "sort: no cards",
  "schedulers_solver_order": 1,
  "schedulers_sigma": "flowmatch",
  "schedulers_prediction_type": "flow_prediction",
  "schedulers_timestep_spacing": "leading",
  "schedulers_beta_schedule": "scaled",
  "cuda_compile_backend": "none",
  "diffusers_to_gpu": true,
  "device_map": "gpu",
  "queue_paused": true
}


Model info


ModuleClassDeviceDTypeParamsModulesConfig
vaeAutoencoderKLxpu:0torch.bfloat16None83819683241

FrozenDict({'in_channels': 3, 'out_channels': 3, 'down_block_types': ['DownEncoderBlock2D', 'DownEncoderBlock2D', 'DownEncoderBlock2D', 'DownEncoderBlock2D'], 'up_block_types': ['UpDecoderBlock2D', 'UpDecoderBlock2D', 'UpDecoderBlock2D', 'UpDecoderBlock2D'], 'block_out_channels': [128, 256, 512, 512], 'layers_per_block': 2, 'act_fn': 'silu', 'latent_channels': 16, 'norm_num_groups': 32, 'sample_size': 1024, 'scaling_factor': 0.3611, 'shift_factor': 0.1159, 'latents_mean': None, 'latents_std': None, 'force_upcast': True, 'use_quant_conv': False, 'use_post_quant_conv': False, 'mid_block_add_attention': True, '_class_name': 'AutoencoderKL', '_diffusers_version': '0.36.0.dev0', '_name_or_path': '/mnt/models/Diffusers/models--vladmandic--flux.1-dev-SRPO/snapshots/690eab4805ea624e20d1abbe4318afba7cccd6b1/vae'})

text_encoderCLIPTextModelxpu:0torch.bfloat16None123060480152

CLIPTextConfig { "architectures": [ "CLIPTextModel" ], "attention_dropout": 0.0, "bos_token_id": 0, "dropout": 0.0, "dtype": "bfloat16", "eos_token_id": 2, "hidden_act": "quick_gelu", "hidden_size": 768, "initializer_factor": 1.0, "initializer_range": 0.02, "intermediate_size": 3072, "layer_norm_eps": 1e-05, "max_position_embeddings": 77, "model_type": "clip_text_model", "num_attention_heads": 12, "num_hidden_layers": 12, "pad_token_id": 1, "projection_dim": 768, "transformers_version": "4.56.1", "vocab_size": 49408 }

text_encoder_2T5EncoderModelxpu:0torch.bfloat16None4762310656463

T5Config { "architectures": [ "T5EncoderModel" ], "classifier_dropout": 0.0, "d_ff": 10240, "d_kv": 64, "d_model": 4096, "decoder_start_token_id": 0, "dense_act_fn": "gelu_new", "dropout_rate": 0.1, "dtype": "bfloat16", "eos_token_id": 1, "feed_forward_proj": "gated-gelu", "initializer_factor": 1.0, "is_encoder_decoder": false, "is_gated_act": true, "layer_norm_epsilon": 1e-06, "model_type": "t5", "num_decoder_layers": 24, "num_heads": 64, "num_layers": 24, "output_past": true, "pad_token_id": 0, "relative_attention_max_distance": 128, "relative_attention_num_buckets": 32, "tie_word_embeddings": false, "transformers_version": "4.56.1", "use_cache": false, "vocab_size": 32128 }

tokenizerCLIPTokenizerNoneNoneNone00

None

tokenizer_2T5TokenizerFastNoneNoneNone00

None

transformerFluxTransformer2DModelxpu:0torch.bfloat16None119014083201279

FrozenDict({'patch_size': 1, 'in_channels': 64, 'out_channels': None, 'num_layers': 19, 'num_single_layers': 38, 'attention_head_dim': 128, 'num_attention_heads': 24, 'joint_attention_dim': 4096, 'pooled_projection_dim': 768, 'guidance_embeds': True, 'axes_dims_rope': [16, 56, 56], '_class_name': 'FluxTransformer2DModel', '_diffusers_version': '0.36.0.dev0', '_name_or_path': 'vladmandic/flux.1-dev-SRPO'})

schedulerFlowMatchEulerDiscreteSchedulerNoneNoneNone00

FrozenDict({'num_train_timesteps': 1000, 'shift': 3.0, 'use_dynamic_shifting': True, 'base_shift': 0.5, 'max_shift': 1.15, 'base_image_seq_len': 256, 'max_image_seq_len': 4096, 'invert_sigmas': False, 'shift_terminal': None, 'use_karras_sigmas': False, 'use_exponential_sigmas': False, 'use_beta_sigmas': False, 'time_shift_type': 'exponential', 'stochastic_sampling': False, '_class_name': 'FlowMatchEulerDiscreteScheduler', '_diffusers_version': '0.36.0.dev0'})

image_encoderNoneTypeNoneNoneNone00

None

feature_extractorNoneTypeNoneNoneNone00

None


  • No labels

Site may be often unavailable due to Russian invasion, but you can Help Ukraine