"'NoneType' object has no attribute 'sd_checkpoint_info'", if prior was "You do not have CLIP state dict!"

Nobody has claimed this yet.

once i get "you don't have state dict", i can't generate an image with the sd model that is set, even if i complete the state dict, due to "'NoneType' object has no attribute 'sd_checkpoint_info'"
only after i change the sd model to something else, generate an image with that one, and then change back to my original sd model, i can generate an image again.

this is the process, that lets me reproduce that issue:

1) start of server

```
Python 3.10.6 (tags/v3.10.6:9c7b4bd, Aug 1 2022, 21:53:49) [MSC v.1932 64 bit (AMD64)]
Version: f2.0.1v1.10.1-previous-584-g9a698e26
Commit hash: 9a698e26d6744de24d05568c9938a52694dbb3f0
Faceswaplab : Use GPU requirements
Checking faceswaplab requirements
Install protobuf>=3.20.2
Installing sd-webui-faceswaplab requirement: protobuf>=3.20.2
1.1965869999985443
CUDA 12.1
Launching Web UI with arguments: --api --port 7861 --gpu-device-id 1 --wildcards-dir 'E:\ai\_gh_repos\sd.webui\webui\extensions\stable-diffusion-webui-wildcards\wildcards' --forge-ref-a1111-home 'E:\ai\_gh_repos\sd.webui\webui' --text-encoder-dir 'E:\ai\_gh_repos\sd.webui\webui\models\text_encoder' --xformers --cuda-malloc --cuda-stream --pin-shared-memory --ckpt-dir 'E:\ai\_gh_repos\sd.webui\webui\models\Stable-diffusion' --vae-dir 'E:\ai\_gh_repos\sd.webui\webui\models\VAE' --hypernetwork-dir 'E:\ai\_gh_repos\sd.webui\webui\models\hypernetworks' --embeddings-dir 'E:\ai\_gh_repos\sd.webui\webui\embeddings' --lora-dir 'E:\ai\_gh_repos\sd.webui\webui\models\lora' --controlnet-dir 'E:\ai\_gh_repos\sd.webui\webui\models\ControlNet' --controlnet-preprocessor-models-dir 'E:\ai\_gh_repos\sd.webui\webui\extensions\sd-webui-controlnet\annotator\downloads'
Set device to: 1
Using cudaMallocAsync backend.
Total VRAM 11264 MB, total RAM 130956 MB
pytorch version: 2.3.1+cu121
xformers version: 0.0.27
Set vram state to: NORMAL_VRAM
Always pin shared GPU memory
Device: cuda:0 NVIDIA GeForce RTX 2080 Ti : cudaMallocAsync
VAE dtype preferences: [torch.float32] -> torch.float32
CUDA Using Stream: True
E:\ai\_gh_repos\webui_forge_cu121_torch231_EXP\system\python\lib\site-packages\transformers\utils\hub.py:127: FutureWarning: Using `TRANSFORMERS_CACHE` is deprecated and will be removed in v5 of Transformers. Use `HF_HOME` instead.
warnings.warn(
Using xformers cross attention
Using xformers attention for VAE
ControlNet preprocessor location: E:\ai\_gh_repos\sd.webui\webui\extensions\sd-webui-controlnet\annotator\downloads
14:29:37 - ReActor - STATUS - Running v0.7.1-b2 on Device: CUDA
Loading additional modules ... done.
2024-10-24 14:29:44,212 - ControlNet - INFO - ControlNet UI callback registered.
Model selected: {'checkpoint_info': {'filename': 'E:\\ai\\_gh_repos\\sd.webui\\webui\\models\\Stable-diffusion\\FLUX\\flux1-dev-bnb-nf4-v2.safetensors', 'hash': 'f0770152'}, 'additional_modules': [], 'unet_storage_dtype': None}
Using online LoRAs in FP16: False
Running on local URL: http://127.0.0.1:7861

To create a public link, set `share=True` in `launch()`.
Startup time: 31.0s (prepare environment: 6.4s, import torch: 7.4s, initialize shared: 0.1s, other imports: 0.4s, load scripts: 2.9s, initialize extra networks: 0.2s, initialize google blockly: 4.4s, create ui: 3.7s, gradio launch: 2.3s, add APIs: 3.1s).
Environment vars changed: {'stream': False, 'inference_memory': 1024.0, 'pin_shared_memory': False}
[GPU Setting] You will use 90.91% GPU memory (10239.00 MB) to load weights, and use 9.09% GPU memory (1024.00 MB) to do matrix computation.
```

2) add vae/encoders one by one:

```
Model selected: {'checkpoint_info': {'filename': 'E:\\ai\\_gh_repos\\sd.webui\\webui\\models\\Stable-diffusion\\FLUX\\flux1-dev-bnb-nf4-v2.safetensors', 'hash': 'f0770152'}, 'additional_modules': ['E:\\ai\\_gh_repos\\sd.webui\\webui\\models\\VAE\\ae.safetensors'], 'unet_storage_dtype': None}
Using online LoRAs in FP16: False
Model selected: {'checkpoint_info': {'filename': 'E:\\ai\\_gh_repos\\sd.webui\\webui\\models\\Stable-diffusion\\FLUX\\flux1-dev-bnb-nf4-v2.safetensors', 'hash': 'f0770152'}, 'additional_modules': ['E:\\ai\\_gh_repos\\sd.webui\\webui\\models\\VAE\\ae.safetensors', 'E:\\ai\\_gh_repos\\sd.webui\\webui\\models\\text_encoder\\clip_l.safetensors'], 'unet_storage_dtype': None}
Using online LoRAs in FP16: False
Model selected: {'checkpoint_info': {'filename': 'E:\\ai\\_gh_repos\\sd.webui\\webui\\models\\Stable-diffusion\\FLUX\\flux1-dev-bnb-nf4-v2.safetensors', 'hash': 'f0770152'}, 'additional_modules': ['E:\\ai\\_gh_repos\\sd.webui\\webui\\models\\VAE\\ae.safetensors', 'E:\\ai\\_gh_repos\\sd.webui\\webui\\models\\text_encoder\\clip_l.safetensors', 'E:\\ai\\_gh_repos\\sd.webui\\webui\\models\\text_encoder\\t5xxl_fp8_e4m3fn.safetensors'], 'unet_storage_dtype': None}
Using online LoRAs in FP16: False
```

3) change flux-model:

```
Model selected: {'checkpoint_info': {'filename': 'E:\\ai\\_gh_repos\\sd.webui\\webui\\models\\Stable-diffusion\\FLUX\\VerusVision_1.0b_Transformer.safetensors', 'hash': '8cb933ca'}, 'additional_modules': ['E:\\ai\\_gh_repos\\sd.webui\\webui\\models\\VAE\\ae.safetensors', 'E:\\ai\\_gh_repos\\sd.webui\\webui\\models\\text_encoder\\clip_l.safetensors', 'E:\\ai\\_gh_repos\\sd.webui\\webui\\models\\text_encoder\\t5xxl_fp8_e4m3fn.safetensors'], 'unet_storage_dtype': None}
Using online LoRAs in FP16: False

```
4) click "generate":

```
Loading Model: {'checkpoint_info': {'filename': 'E:\\ai\\_gh_repos\\sd.webui\\webui\\models\\Stable-diffusion\\FLUX\\VerusVision_1.0b_Transformer.safetensors', 'hash': '8cb933ca'}, 'additional_modules': ['E:\\ai\\_gh_repos\\sd.webui\\webui\\models\\VAE\\ae.safetensors', 'E:\\ai\\_gh_repos\\sd.webui\\webui\\models\\text_encoder\\clip_l.safetensors', 'E:\\ai\\_gh_repos\\sd.webui\\webui\\models\\text_encoder\\t5xxl_fp8_e4m3fn.safetensors'], 'unet_storage_dtype': None}
[Unload] Trying to free all memory for cuda:0 with 0 models keep loaded ... Done.
StateDict Keys: {'transformer': 776, 'vae': 244, 'text_encoder': 196, 'text_encoder_2': 220, 'ignore': 0}
Using Detected T5 Data Type: torch.float8_e4m3fn
Using Detected UNet Type: torch.float8_e4m3fn
Working with z of shape (1, 16, 32, 32) = 16384 dimensions.
K-Model Created: {'storage_dtype': torch.float8_e4m3fn, 'computation_dtype': torch.float16}
Model loaded in 1.2s (unload existing model: 0.4s, forge model load: 0.7s).
Warning: field infotext in API payload not found in .
[Unload] Trying to free 7723.54 MB for cuda:0 with 0 models keep loaded ... Done.
[Memory Management] Target: JointTextEncoder, Free GPU: 10085.31 MB, Model Require: 5153.49 MB, Previously Loaded: 0.00 MB, Inference Require: 1024.00 MB, Remaining: 3907.81 MB, All loaded to GPU.
Moving model(s) has taken 2.43 seconds
Distilled CFG Scale will be ignored for Schnell
[Unload] Trying to free 1024.00 MB for cuda:0 with 1 models keep loaded ... Current free memory is 4808.97 MB ... Done.
Distilled CFG Scale will be ignored for Schnell
[Unload] Trying to free 16032.65 MB for cuda:0 with 0 models keep loaded ... Current free memory is 4803.98 MB ... Unload model JointTextEncoder Done.
[Memory Management] Target: KModel, Free GPU: 10029.96 MB, Model Require: 11340.31 MB, Previously Loaded: 0.00 MB, Inference Require: 1024.00 MB, Remaining: -2334.35 MB, CPU Swap Loaded (blocked method): 3636.00 MB, GPU Loaded: 7704.31 MB
Moving model(s) has taken 5.14 seconds
100%|#######################################################################################################################################################################################################################################################################################| 20/20 [01:28<00:00, 4.41s/it]
[Unload] Trying to free 8991.55 MB for cuda:0 with 0 models keep loaded ... Current free memory is 2305.54 MB ... Unload model KModel Done.█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 20/20 [01:22<00:00, 4.48s/it]
[Memory Management] Target: IntegratedAutoencoderKL, Free GPU: 10009.87 MB, Model Require: 319.75 MB, Previously Loaded: 0.00 MB, Inference Require: 1024.00 MB, Remaining: 8666.12 MB, All loaded to GPU.
Moving model(s) has taken 2.79 seconds
Total progress: 100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 20/20 [01:27<00:00, 4.36s/it]
```

5) delete vae/encoders one by one in order to force "you don't have state dict":

```
Model selected: {'checkpoint_info': {'filename': 'E:\\ai\\_gh_repos\\sd.webui\\webui\\models\\Stable-diffusion\\FLUX\\VerusVision_1.0b_Transformer.safetensors', 'hash': '8cb933ca'}, 'additional_modules': ['E:\\ai\\_gh_repos\\sd.webui\\webui\\models\\text_encoder\\clip_l.safetensors', 'E:\\ai\\_gh_repos\\sd.webui\\webui\\models\\text_encoder\\t5xxl_fp8_e4m3fn.safetensors'], 'unet_storage_dtype': None}
Using online LoRAs in FP16: False
Model selected: {'checkpoint_info': {'filename': 'E:\\ai\\_gh_repos\\sd.webui\\webui\\models\\Stable-diffusion\\FLUX\\VerusVision_1.0b_Transformer.safetensors', 'hash': '8cb933ca'}, 'additional_modules': ['E:\\ai\\_gh_repos\\sd.webui\\webui\\models\\text_encoder\\t5xxl_fp8_e4m3fn.safetensors'], 'unet_storage_dtype': None}
Using online LoRAs in FP16: False
Model selected: {'checkpoint_info': {'filename': 'E:\\ai\\_gh_repos\\sd.webui\\webui\\models\\Stable-diffusion\\FLUX\\VerusVision_1.0b_Transformer.safetensors', 'hash': '8cb933ca'}, 'additional_modules': [], 'unet_storage_dtype': None}
Using online LoRAs in FP16: False
```

6) click "generate":

```
Loading Model: {'checkpoint_info': {'filename': 'E:\\ai\\_gh_repos\\sd.webui\\webui\\models\\Stable-diffusion\\FLUX\\VerusVision_1.0b_Transformer.safetensors', 'hash': '8cb933ca'}, 'additional_modules': [], 'unet_storage_dtype': None}
[Unload] Trying to free all memory for cuda:0 with 0 models keep loaded ... Current free memory is 9681.11 MB ... Unload model IntegratedAutoencoderKL Done.
StateDict Keys: {'transformer': 776, 'vae': 0, 'ignore': 0}
Traceback (most recent call last):
File "E:\ai\_gh_repos\webui_forge_cu121_torch231_EXP\webui\modules_forge\main_thread.py", line 30, in work
self.result = self.func(*self.args, **self.kwargs)
File "E:\ai\_gh_repos\webui_forge_cu121_torch231_EXP\webui\modules\txt2img.py", line 125, in txt2img_function
processed = processing.process_images(p)
File "E:\ai\_gh_repos\webui_forge_cu121_torch231_EXP\webui\modules\processing.py", line 834, in process_images
manage_model_and_prompt_cache(p)
File "E:\ai\_gh_repos\webui_forge_cu121_torch231_EXP\webui\modules\processing.py", line 802, in manage_model_and_prompt_cache
p.sd_model, just_reloaded = forge_model_reload()
File "E:\ai\_gh_repos\webui_forge_cu121_torch231_EXP\system\python\lib\site-packages\torch\utils\_contextlib.py", line 115, in decorate_context
return func(*args, **kwargs)
File "E:\ai\_gh_repos\webui_forge_cu121_torch231_EXP\webui\modules\sd_models.py", line 504, in forge_model_reload
sd_model = forge_loader(state_dict, additional_state_dicts=additional_state_dicts)
File "E:\ai\_gh_repos\webui_forge_cu121_torch231_EXP\system\python\lib\site-packages\torch\utils\_contextlib.py", line 115, in decorate_context
return func(*args, **kwargs)
File "E:\ai\_gh_repos\webui_forge_cu121_torch231_EXP\webui\backend\loader.py", line 285, in forge_loader
component = load_huggingface_component(estimated_config, component_name, lib_name, cls_name, local_path, component_sd)
File "E:\ai\_gh_repos\webui_forge_cu121_torch231_EXP\webui\backend\loader.py", line 59, in load_huggingface_component
assert isinstance(state_dict, dict) and len(state_dict) > 16, 'You do not have CLIP state dict!'
AssertionError: You do not have CLIP state dict!
You do not have CLIP state dict!
```

7) add back vae/encoders one by one:

```
Model selected: {'checkpoint_info': {'filename': 'E:\\ai\\_gh_repos\\sd.webui\\webui\\models\\Stable-diffusion\\FLUX\\VerusVision_1.0b_Transformer.safetensors', 'hash': '8cb933ca'}, 'additional_modules': ['E:\\ai\\_gh_repos\\sd.webui\\webui\\models\\VAE\\ae.safetensors'], 'unet_storage_dtype': None}
Using online LoRAs in FP16: False
Model selected: {'checkpoint_info': {'filename': 'E:\\ai\\_gh_repos\\sd.webui\\webui\\models\\Stable-diffusion\\FLUX\\VerusVision_1.0b_Transformer.safetensors', 'hash': '8cb933ca'}, 'additional_modules': ['E:\\ai\\_gh_repos\\sd.webui\\webui\\models\\VAE\\ae.safetensors', 'E:\\ai\\_gh_repos\\sd.webui\\webui\\models\\text_encoder\\clip_l.safetensors'], 'unet_storage_dtype': None}
Using online LoRAs in FP16: False
Model selected: {'checkpoint_info': {'filename': 'E:\\ai\\_gh_repos\\sd.webui\\webui\\models\\Stable-diffusion\\FLUX\\VerusVision_1.0b_Transformer.safetensors', 'hash': '8cb933ca'}, 'additional_modules': ['E:\\ai\\_gh_repos\\sd.webui\\webui\\models\\VAE\\ae.safetensors', 'E:\\ai\\_gh_repos\\sd.webui\\webui\\models\\text_encoder\\clip_l.safetensors', 'E:\\ai\\_gh_repos\\sd.webui\\webui\\models\\text_encoder\\t5xxl_fp8_e4m3fn.safetensors'], 'unet_storage_dtype': None}
Using online LoRAs in FP16: False
```

8) click "generate":

```
[Unload] Trying to free all memory for cuda:0 with 0 models keep loaded ... Done.
Traceback (most recent call last):
File "E:\ai\_gh_repos\webui_forge_cu121_torch231_EXP\webui\modules_forge\main_thread.py", line 30, in work
self.result = self.func(*self.args, **self.kwargs)
File "E:\ai\_gh_repos\webui_forge_cu121_torch231_EXP\webui\modules\txt2img.py", line 125, in txt2img_function
processed = processing.process_images(p)
File "E:\ai\_gh_repos\webui_forge_cu121_torch231_EXP\webui\modules\processing.py", line 840, in process_images
res = process_images_inner(p)
File "E:\ai\_gh_repos\webui_forge_cu121_torch231_EXP\webui\modules\processing.py", line 877, in process_images_inner
p.sd_model_name = shared.sd_model.sd_checkpoint_info.name_for_extra
AttributeError: 'NoneType' object has no attribute 'sd_checkpoint_info'
'NoneType' object has no attribute 'sd_checkpoint_info'
```

Reproduce the missing-state-dict sequence, then trace the failure from modules/processing.py and modules/sd_models.py into backend/loader.py, where the traceback reaches load_huggingface_component. Compare model reload behavior before and after the VAE and text encoders are restored. Done means the selected model can generate again without temporarily switching to another model.

Written by the indexing model from the issue text.