invoke-ai
[bug]: Crash using Flux Dev.1
Is there an existing issue for this problem?
- [X] I have searched the existing issues
Operating system
Windows
GPU vendor
Nvidia (CUDA)
GPU model
3060
GPU VRAM
12
Version number
5.5.0
Browser
Edge
Python dependencies
Local System accelerate
1.0.1
compel
2.0.2
cuda
12.4
diffusers
0.31.0
numpy
1.26.3
opencv
4.9.0.80
onnx
1.16.1
pillow
10.2.0
python
3.11.11
torch
2.4.1+cu124
torchvision
0.19.1+cu124
transformers
4.46.3
xformers
Not Installed
What happened
click generate and it exits with: [2025-01-06 18:09:22,644]::[InvokeAI]::INFO --> Cleaned database (freed 0.01MB) [2025-01-06 18:09:22,644]::[InvokeAI]::INFO --> Invoke running on http://0.0.0.0:9090/ (Press CTRL+C to quit) [2025-01-06 18:11:10,050]::[InvokeAI]::INFO --> Executing queue item 91, session 1d1a923c-b431-4d28-b25e-0f1bc38690fa C:\InvokeAI.venv\Lib\site-packages\bitsandbytes\autograd_functions.py:316: UserWarning: MatMul8bitLt: inputs will be cast from torch.bfloat16 to float16 during quantization warnings.warn(f"MatMul8bitLt: inputs will be cast from {A.dtype} to float16 during quantization") C:\InvokeAI.venv\Lib\site-packages\transformers\models\clip\modeling_clip.py:540: UserWarning: 1Torch was not compiled with flash attention. (Triggered internally at C:\actions-runner_work\pytorch\pytorch\builder\windows\pytorch\aten\src\ATen\native\transformers\cuda\sdp_utils.cpp:555.) attn_output = torch.nn.functional.scaled_dot_product_attention( Process exited with code: 3221225477
What you expected to happen
on generation it loads and then exits within about 20 seconds
How to reproduce the problem
generate anything with the Flux DEV non quant mode
Additional context
TY I am very new to this but trying hard to learn
Discord username
No response
I'm getting similar with Flux Schnell on a 2070 Super with Invoke installed viaStability Matrix:
\StabilityMatrix\Packages\InvokeAI\venv\lib\site-packages\transformers\models\clip\modeling_clip.py:540: UserWarning: 1Torch was not compiled with flash attention. (Triggered internally at C:\actions-runner\_work\pytorch\pytorch\builder\windows\pytorch\aten\src\ATen\native\transformers\cuda\sdp_utils.cpp:555.) attn_output = torch.nn.functional.scaled_dot_product_attention(
Same here.... started happening a few days ago. No idea what caused it. So annoying! Anyone know a fix please?
I get this:
[2025-01-12 03:08:45,916]::[ModelInstallService]::INFO --> Model download complete: black-forest-labs/FLUX.1-dev [2025-01-12 03:08:45,920]::[ModelInstallService]::INFO --> Model install started: black-forest-labs/FLUX.1-dev [2025-01-12 03:08:45,924]::[ModelInstallService]::ERROR --> Model install error: black-forest-labs/FLUX.1-dev InvalidModelConfigException: Unknown base model for /invokeai/models/tmpinstall_fvh2hdgc/FLUX.1-dev
@gigend @KudintG @tensorflow73 Just to confirm, are you all seeing Process exited with code: 3221225477? Or just the same warnings that lead up to it? And, can you all confirm what version of Invoke you are seeing this on?
@TiddlyWiddly your error is unrelated to the main issue here. Please open a new bug report.
I've found that the base Flux schnell model with the standard T5 works fine, but other flux models crash it out with the error pointing to the flash attention build issue. Pony, sd1.x and sdxl models all function normally.
I've found that the base Flux schnell model with the standard T5 works fine, but other flux models crash it out with the error pointing to the flash attention build issue. Pony, sd1.x and sdxl models all function normally.
The warnings about flash attention are expected on Windows. The real error in the original bug report is Process exited with code: 3221225477. Are you seeing this same error? Or something else? And, what version of Invoke are you running?
I am seeing the same error in 5.6.0:
Starting up...
Started Invoke process with PID: 24132
The installed version of bitsandbytes was compiled without GPU support. 8-bit optimizers, 8-bit multiplication, and GPU quantization are unavailable.
[2025-01-26 15:17:13,641]::[InvokeAI]::INFO --> Loading node pack clipInterrogator-invokeai-node
[2025-01-26 15:17:14,468]::[InvokeAI]::INFO --> Loaded 1 node packs from C:\Users\joe\invokeai2\nodes
[2025-01-26 15:17:17,605]::[InvokeAI]::INFO --> Patchmatch initialized
[2025-01-26 15:17:19,427]::[InvokeAI]::INFO --> Using torch device: CPU
[2025-01-26 15:17:20,909]::[InvokeAI]::INFO --> InvokeAI version 5.6.0
[2025-01-26 15:17:20,910]::[InvokeAI]::INFO --> Root directory = C:\Users\joe\invokeai2
[2025-01-26 15:17:20,915]::[InvokeAI]::INFO --> Initializing database at C:\Users\joe\invokeai2\databases\invokeai.db
[2025-01-26 15:17:20,943]::[ModelManagerService]::INFO --> [MODEL CACHE] Calculated model RAM cache size: 5599.30 MB. Heuristics applied: [1].
[2025-01-26 15:17:21,008]::[InvokeAI]::INFO --> Pruned 1 finished queue items
[2025-01-26 15:17:21,382]::[InvokeAI]::INFO --> Cleaned database (freed 0.01MB)
[2025-01-26 15:17:21,382]::[InvokeAI]::INFO --> Invoke running on http://127.0.0.1:9090/ (Press CTRL+C to quit)
[2025-01-26 15:17:21,406]::[InvokeAI]::INFO --> Executing queue item 6581, session b4735278-a291-444d-a7da-b1b25edcdafa
Process exited with code 3221225477
See the docs here for tips on tuning memory usage for your system: https://invoke-ai.github.io/InvokeAI/features/low-vram
In particular, this section addresses common causes of exit code 3221225477: https://invoke-ai.github.io/InvokeAI/features/low-vram/?h=low+vram#windows-page-file
Same issue here.
I have a GeForce RTX 4070 TI and i'm trying to install the "FLUX.1 Kontext dev" model on Invoke v6.0.0rc3.
Log:
[2025-07-05 02:03:19,554]::[ModelInstallService]::INFO --> Queueing model install: black-forest-labs/FLUX.1-Kontext-dev::flux1-kontext-dev.safetensors (1 file)
[2025-07-05 02:03:19,555]::[InvokeAI]::INFO --> Started installation of black-forest-labs/FLUX.1-Kontext-dev::flux1-kontext-dev.safetensors
[2025-07-05 02:03:20,356]::[DownloadQueueService]::INFO --> File download started: https://huggingface.co/black-forest-labs/FLUX.1-Kontext-dev/resolve/main/flux1-kontext-dev.safetensors
[2025-07-05 02:07:03,038]::[DownloadQueueService]::INFO --> Download complete: https://huggingface.co/black-forest-labs/FLUX.1-Kontext-dev/resolve/main/flux1-kontext-dev.safetensors
[2025-07-05 02:07:03,038]::[ModelInstallService]::INFO --> Model download complete: black-forest-labs/FLUX.1-Kontext-dev::flux1-kontext-dev.safetensors
[2025-07-05 02:07:03,039]::[ModelInstallService]::INFO --> Model install started: black-forest-labs/FLUX.1-Kontext-dev::flux1-kontext-dev.safetensors
Hashing FLUX.1-Kontext-dev_flux1-kontext-dev.safetensors: 0%| | 0/1 [00:00<?, ?file/s]
Hashing FLUX.1-Kontext-dev_flux1-kontext-dev.safetensors: 0%| | 0/1 [00:00<?, ?file/s]
Hashing FLUX.1-Kontext-dev_flux1-kontext-dev.safetensors: 100%|##########| 1/1 [01:06<00:00, 66.12s/file]
Hashing FLUX.1-Kontext-dev_flux1-kontext-dev.safetensors: 100%|##########| 1/1 [01:06<00:00, 66.13s/file]
Process exited with code 3221225477
It turned out my problem was related to low disk space. I've installed invoke on a secondary ssd with more than 80GB of free space but during the installation of the model it uses my primary disk (that only had 20GB free) for paging. The official invoke documentation helped me to find the solution.
https://invoke-ai.github.io/InvokeAI/features/low-vram/#windows-page-file
Same problem here with v1.8.1, Windows 11, RTX 4090 (24gb), 64gb RAM, >1TB free disk space, "system managed" page size. hashing_algorithm: random has no effect. Using "in-place install" works for all but FLUX.1-Fill-dev_flux1-fill-dev.safetensors which still fails with Process exited with code 3221225477
