[bug]: Unable to Load FLUX Models in InvokeAI

[greyrabbit2003]

Is there an existing issue for this problem?

• I have searched the existing issues

Operating system

Linux

GPU vendor

Nvidia (CUDA)

GPU model

RTX 3060

GPU VRAM

12

Version number

v5.7.1

Browser

Firefox

Python dependencies

I installed InvokeAI through Pinokio Browser, and while SD and SDXL models work fine, FLUX models fail to load. The error suggests an issue with model loading, specifically referencing stable-diffusion/v1-inference.yaml.

Steps to Reproduce:

1. Installed InvokeAI via Pinokio Browser.
   

2. Loaded SD and SDXL models successfully.
   

3. Tried loading FLLLLLLUX models, but they fail.
   

4. Error appears in logs (attached below).
   

`<<PINOKIO_SHELL>>eval "$(conda shell.bash hook)" ; conda deactivate ; conda deactivate ; conda deactivate ; conda activate base ; source /home/ubuntu/pinokio/api/invoke.git/app/env/bin/activate /home/ubuntu/pinokio/api/invoke.git/app/env ; invokeai-web

patchmatch.patch_match: INFO - Compiling and loading c extensions from "/home/ubuntu/pinokio/api/invoke.git/app/env/lib/python3.10/site-packages/patchmatch".
patchmatch.patch_match: ERROR - patchmatch failed to load or compile (Command 'make clean && make' returned non-zero exit status 2.).
patchmatch.patch_match: INFO - Refer to https://invoke-ai.github.io/InvokeAI/installation/060_INSTALL_PATCHMATCH/ for installation instructions.
[2025-03-05 20:07:02,024]::[InvokeAI]::INFO --> Patchmatch not loaded (nonfatal)
[2025-03-05 20:07:02,696]::[InvokeAI]::INFO --> Using torch device: NVIDIA GeForce RTX 3060
[2025-03-05 20:07:03,013]::[InvokeAI]::INFO --> cuDNN version: 90100
[2025-03-05 20:07:03,022]::[InvokeAI]::INFO --> InvokeAI version 5.7.1
[2025-03-05 20:07:03,022]::[InvokeAI]::INFO --> Root directory = /home/ubuntu/pinokio/api/invoke.git/app
[2025-03-05 20:07:03,023]::[InvokeAI]::INFO --> Initializing database at /home/ubuntu/pinokio/drive/drives/peers/d1741151917640/databases/invokeai.db
[2025-03-05 20:07:03,125]::[ModelManagerService]::INFO --> [MODEL CACHE] Calculated model RAM cache size: 8849.00 MB. Heuristics applied: [1, 2].
[2025-03-05 20:07:03,144]::[ModelInstallService]::WARNING --> Missing model file: Shakker-Labs-FLUX.1-dev-ControlNet-Union-Pro at /home/ubuntu/pinokio/drive/drives/peers/d1739623879987/controlnet/Shakker-Labs-FLUX.1-dev-ControlNet-Union-Pro.safetensors
[2025-03-05 20:07:03,144]::[ModelInstallService]::WARNING --> Missing model file: hunyuan_720_vae_t2v_fp8 at /home/ubuntu/pinokio/drive/drives/peers/d1739623879987/vae/hunyuan_720_vae_t2v_fp8.pt
[2025-03-05 20:07:03,144]::[ModelInstallService]::WARNING --> Missing model file: hunyuan_video_vae_bf16 at /home/ubuntu/pinokio/drive/drives/peers/d1739623879987/vae/hunyuan_video_vae_bf16.safetensors
[2025-03-05 20:07:03,144]::[ModelInstallService]::WARNING --> Missing model file: vae-THUDM-CogVideoX-2b at /home/ubuntu/pinokio/drive/drives/peers/d1739623879987/vae/vae-THUDM-CogVideoX-2b.safetensors
[2025-03-05 20:07:03,144]::[ModelInstallService]::WARNING --> Missing model file: vae_Lightricks-LTX-Video at /home/ubuntu/pinokio/drive/drives/peers/d1739623879987/vae/vae_Lightricks-LTX-Video.safetensors
[2025-03-05 20:07:03,282]::[InvokeAI]::INFO --> Invoke running on http://127.0.0.1:9090 (Press CTRL+C to quit)
[2025-03-05 20:07:48,878]::[InvokeAI]::INFO --> Deleted model: 8bb5e3d3-cb56-4d36-8262-bddff6f6fba6
[2025-03-05 20:07:51,749]::[InvokeAI]::INFO --> Deleted model: dfc76c0f-ba2c-4f10-bd3d-4052e8066e91
[2025-03-05 20:07:54,482]::[InvokeAI]::INFO --> Deleted model: 109ac880-3fed-4587-a3f8-6b0c9a1fbf79
[2025-03-05 20:08:05,773]::[InvokeAI]::INFO --> Deleted model: 67ef1e87-2e2d-4ba2-b993-b509c8e225f1
[2025-03-05 20:08:15,377]::[InvokeAI]::INFO --> Updated model: 3376cdc0-e63e-401e-9cf0-696a857589d4
[2025-03-05 20:08:27,199]::[InvokeAI]::INFO --> Updated model: 4485a72c-9c71-4af0-8156-8842cc84408a
[2025-03-05 20:08:54,745]::[InvokeAI]::INFO --> Deleted model: 6be625a1-039f-4818-9d71-a120d00aee49
[2025-03-05 20:09:16,417]::[InvokeAI]::INFO --> Started installation of /home/ubuntu/pinokio/drive/drives/peers/d1739623879987/controlnet/FLUX/Shakker-Labs-FLUX.1-dev-ControlNet-Union-Pro.safetensors
[2025-03-05 20:09:16,417]::[ModelInstallService]::INFO --> Model install started: /home/ubuntu/pinokio/drive/drives/peers/d1739623879987/controlnet/FLUX/Shakker-Labs-FLUX.1-dev-ControlNet-Union-Pro.safetensors
Hashing Shakker-Labs-FLUX.1-dev-ControlNet-Union-Pro.safetensors: 100%|██████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 1/1 [00:06<00:00, 6.56s/file]
[2025-03-05 20:09:23,251]::[ModelInstallService]::INFO --> Model install complete: /home/ubuntu/pinokio/drive/drives/peers/d1739623879987/controlnet/FLUX/Shakker-Labs-FLUX.1-dev-ControlNet-Union-Pro.safetensors
[2025-03-05 20:09:58,381]::[InvokeAI]::INFO --> Executing queue item 225, session c1746ee2-9901-4cf0-9786-946e5fa50d09
Loading checkpoint shards: 100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 2/2 [00:00<00:00, 2.97it/s]
[2025-03-05 20:10:14,123]::[ModelManagerService]::INFO --> [MODEL CACHE] Loaded model '57198c6b-794c-42ce-aa90-bf32f081ff17:text_encoder_2' (T5EncoderModel) onto cuda device in 14.77s. Total model size: 9083.39MB, VRAM: 9083.39MB (100.0%)
[2025-03-05 20:10:14,235]::[ModelManagerService]::INFO --> [MODEL CACHE] Loaded model '57198c6b-794c-42ce-aa90-bf32f081ff17:tokenizer_2' (T5Tokenizer) onto cuda device in 0.00s. Total model size: 0.03MB, VRAM: 0.00MB (0.0%)
[2025-03-05 20:10:15,282]::[ModelManagerService]::INFO --> [MODEL CACHE] Loaded model 'c1b60061-0709-4367-ba7a-03a88ec2f770:text_encoder' (CLIPTextModel) onto cuda device in 0.07s. Total model size: 469.44MB, VRAM: 469.44MB (100.0%)
[2025-03-05 20:10:15,341]::[ModelManagerService]::INFO --> [MODEL CACHE] Loaded model 'c1b60061-0709-4367-ba7a-03a88ec2f770:tokenizer' (CLIPTokenizer) onto cuda device in 0.00s. Total model size: 0.00MB, VRAM: 0.00MB (0.0%)
[2025-03-05 20:10:16,479]::[ModelManagerService]::INFO --> [MODEL CACHE] Loaded model '600dcc16-38ba-4a3b-87b7-7054c6aac570:transformer' (Flux) onto cuda device in 0.95s. Total model size: 5679.45MB, VRAM: 5679.45MB (100.0%)
33%|█████████████████████████████████████████████████████████████▎ | 10/30 [00:39<01:19, 3.99s/it]
100%|████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 30/30 [01:59<00:00, 3.99s/it]
[2025-03-05 20:12:16,556]::[InvokeAI]::ERROR --> Error while invoking session c1746ee2-9901-4cf0-9786-946e5fa50d09, invocation d811d337-44c7-4afd-8fea-84a3027df098 (flux_vae_decode): 'stable-diffusion/v1-inference.yaml'
[2025-03-05 20:12:16,556]::[InvokeAI]::ERROR --> Traceback (most recent call last):
File "/home/ubuntu/pinokio/api/invoke.git/app/env/lib/python3.10/site-packages/invokeai/app/services/session_processor/session_processor_default.py", line 129, in run_node
output = invocation.invoke_internal(context=context, services=self._services)
File "/home/ubuntu/pinokio/api/invoke.git/app/env/lib/python3.10/site-packages/invokeai/app/invocations/baseinvocation.py", line 303, in invoke_internal
output = self.invoke(context)
File "/home/ubuntu/pinokio/api/invoke.git/app/env/lib/python3.10/site-packages/torch/utils/_contextlib.py", line 116, in decorate_context
return func(*args, **kwargs)
File "/home/ubuntu/pinokio/api/invoke.git/app/env/lib/python3.10/site-packages/invokeai/app/invocations/flux_vae_decode.py", line 72, in invoke
vae_info = context.models.load(self.vae.vae)
File "/home/ubuntu/pinokio/api/invoke.git/app/env/lib/python3.10/site-packages/invokeai/app/services/shared/invocation_context.py", line 397, in load
return self._services.model_manager.load.load_model(model, submodel_type)
File "/home/ubuntu/pinokio/api/invoke.git/app/env/lib/python3.10/site-packages/invokeai/app/services/model_load/model_load_default.py", line 70, in load_model
).load_model(model_config, submodel_type)
File "/home/ubuntu/pinokio/api/invoke.git/app/env/lib/python3.10/site-packages/invokeai/backend/model_manager/load/load_default.py", line 58, in load_model
cache_record = self._load_and_cache(model_config, submodel_type)
File "/home/ubuntu/pinokio/api/invoke.git/app/env/lib/python3.10/site-packages/invokeai/backend/model_manager/load/load_default.py", line 79, in _load_and_cache
loaded_model = self._load_model(config, submodel_type)
File "/home/ubuntu/pinokio/api/invoke.git/app/env/lib/python3.10/site-packages/invokeai/backend/model_manager/load/model_loaders/flux.py", line 84, in _load_model
model = AutoEncoder(ae_params[config.config_path])
KeyError: 'stable-diffusion/v1-inference.yaml'

[2025-03-05 20:12:16,606]::[InvokeAI]::INFO --> Graph stats: c1746ee2-9901-4cf0-9786-946e5fa50d09
Node Calls Seconds VRAM Used
flux_model_loader 1 0.013s 0.000G
lora_selector 1 0.002s 0.000G
collect 2 0.002s 0.472G
flux_lora_collection_loader 1 0.001s 0.000G
flux_text_encoder 1 16.994s 9.118G
flux_denoise 1 121.123s 7.427G
flux_vae_decode 1 0.022s 6.844G
TOTAL GRAPH EXECUTION TIME: 138.157s
TOTAL GRAPH WALL TIME: 138.161s
RAM used by InvokeAI process: 8.31G (+7.398G)
RAM used to load models: 14.89G
VRAM in use: 6.844G
RAM cache statistics:
Model cache hits: 8
Model cache misses: 7
Models cached: 5
Models cleared from cache: 1
Cache high water mark: 8.89/0.00G

[2025-03-05 20:12:27,772]::[InvokeAI]::INFO --> Executing queue item 226, session f4820122-acc7-4e19-94ee-a395020de121
Loading checkpoint shards: 100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 2/2 [00:00<00:00, 5.68it/s]
[2025-03-05 20:12:30,154]::[ModelManagerService]::INFO --> [MODEL CACHE] Loaded model '57198c6b-794c-42ce-aa90-bf32f081ff17:text_encoder_2' (T5EncoderModel) onto cuda device in 1.22s. Total model size: 9083.39MB, VRAM: 9083.39MB (100.0%)
[2025-03-05 20:12:30,251]::[ModelManagerService]::INFO --> [MODEL CACHE] Loaded model '57198c6b-794c-42ce-aa90-bf32f081ff17:tokenizer_2' (T5Tokenizer) onto cuda device in 0.00s. Total model size: 0.03MB, VRAM: 0.00MB (0.0%)
[2025-03-05 20:12:33,006]::[ModelManagerService]::INFO --> [MODEL CACHE] Loaded model 'c1b60061-0709-4367-ba7a-03a88ec2f770:text_encoder' (CLIPTextModel) onto cuda device in 0.09s. Total model size: 469.44MB, VRAM: 469.44MB (100.0%)
[2025-03-05 20:12:33,059]::[ModelManagerService]::INFO --> [MODEL CACHE] Loaded model 'c1b60061-0709-4367-ba7a-03a88ec2f770:tokenizer' (CLIPTokenizer) onto cuda device in 0.00s. Total model size: 0.00MB, VRAM: 0.00MB (0.0%)
[2025-03-05 20:12:34,459]::[ModelManagerService]::INFO --> [MODEL CACHE] Loaded model '600dcc16-38ba-4a3b-87b7-7054c6aac570:transformer' (Flux) onto cuda device in 1.20s. Total model size: 5679.45MB, VRAM: 5679.45MB (100.0%)
100%|████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 30/30 [02:00<00:00, 4.01s/it]
[2025-03-05 20:14:35,039]::[InvokeAI]::ERROR --> Error while invoking session f4820122-acc7-4e19-94ee-a395020de121, invocation 14db28d5-75ce-40fa-808c-72d58a63be19 (flux_vae_decode): 'stable-diffusion/v1-inference.yaml'
[2025-03-05 20:14:35,039]::[InvokeAI]::ERROR --> Traceback (most recent call last):
File "/home/ubuntu/pinokio/api/invoke.git/app/env/lib/python3.10/site-packages/invokeai/app/services/session_processor/session_processor_default.py", line 129, in run_node
output = invocation.invoke_internal(context=context, services=self._services)
File "/home/ubuntu/pinokio/api/invoke.git/app/env/lib/python3.10/site-packages/invokeai/app/invocations/baseinvocation.py", line 303, in invoke_internal
output = self.invoke(context)
File "/home/ubuntu/pinokio/api/invoke.git/app/env/lib/python3.10/site-packages/torch/utils/_contextlib.py", line 116, in decorate_context
return func(*args, **kwargs)
File "/home/ubuntu/pinokio/api/invoke.git/app/env/lib/python3.10/site-packages/invokeai/app/invocations/flux_vae_decode.py", line 72, in invoke
vae_info = context.models.load(self.vae.vae)
File "/home/ubuntu/pinokio/api/invoke.git/app/env/lib/python3.10/site-packages/invokeai/app/services/shared/invocation_context.py", line 397, in load
return self._services.model_manager.load.load_model(model, submodel_type)
File "/home/ubuntu/pinokio/api/invoke.git/app/env/lib/python3.10/site-packages/invokeai/app/services/model_load/model_load_default.py", line 70, in load_model
).load_model(model_config, submodel_type)
File "/home/ubuntu/pinokio/api/invoke.git/app/env/lib/python3.10/site-packages/invokeai/backend/model_manager/load/load_default.py", line 58, in load_model
cache_record = self._load_and_cache(model_config, submodel_type)
File "/home/ubuntu/pinokio/api/invoke.git/app/env/lib/python3.10/site-packages/invokeai/backend/model_manager/load/load_default.py", line 79, in _load_and_cache
loaded_model = self._load_model(config, submodel_type)
File "/home/ubuntu/pinokio/api/invoke.git/app/env/lib/python3.10/site-packages/invokeai/backend/model_manager/load/model_loaders/flux.py", line 84, in _load_model
model = AutoEncoder(ae_params[config.config_path])
KeyError: 'stable-diffusion/v1-inference.yaml'

[2025-03-05 20:14:35,093]::[InvokeAI]::INFO --> Graph stats: f4820122-acc7-4e19-94ee-a395020de121
Node Calls Seconds VRAM Used
flux_model_loader 1 0.001s 6.844G
lora_selector 1 0.000s 6.844G
collect 2 0.001s 6.844G
flux_lora_collection_loader 1 0.002s 6.844G
flux_text_encoder 1 5.305s 9.129G
flux_denoise 1 121.917s 7.432G
flux_vae_decode 1 0.025s 6.850G
TOTAL GRAPH EXECUTION TIME: 127.251s
TOTAL GRAPH WALL TIME: 127.259s
RAM used by InvokeAI process: 8.46G (+0.148G)
RAM used to load models: 14.89G
VRAM in use: 6.850G
RAM cache statistics:
Model cache hits: 8
Model cache misses: 7
Models cached: 5
Models cleared from cache: 1
Cache high water mark: 8.89/0.00G
`

What happened

he model fails to load with an error mentioning stable-diffusion/v1-inference.yaml.

What you expected to happen

FLUX models should load successfully like SD and SDXL models.

How to reproduce the problem

No response

Additional context

No response

Discord username

No response

[psyins]

I have the exact same problem. It was working fine until yesterday, but suddenly this error occurred.
I don't know where the problem is coming from. Does anyone have a solution?