[GH-ISSUE #152] "Illegal instruction" on RPI 4b #86

New issue

Closed

opened 2026-10-15 18:04:17 +00:00 by bsakaguchi · 5 comments

bsakaguchi commented

2026-10-15 18:04:17 +00:00

Owner

Originally created by @NyakoFox on GitHub (Sep 9, 2023).
Original GitHub issue: https://github.com/SillyTavern/SillyTavern-Extras/issues/152

OS: Raspberry Pi OS (64-bit)
Python: 3.11.5
Using miniforge instead of Miniconda as it doesn't seem to work on RPI?

When trying to run SillyTavern-Extras, I get "Illegal instruction" with no further output. By doing some debug prints, it seems to be on the line from transformers import AutoTokenizer, AutoProcessor, pipeline

Originally created by @NyakoFox on GitHub (Sep 9, 2023). Original GitHub issue: https://github.com/SillyTavern/SillyTavern-Extras/issues/152 - OS: Raspberry Pi OS (64-bit) - Python: 3.11.5 - Using [miniforge](https://github.com/conda-forge/miniforge) instead of Miniconda as it doesn't seem to work on RPI? When trying to run SillyTavern-Extras, I get "Illegal instruction" with no further output. By doing some debug prints, it seems to be on the line `from transformers import AutoTokenizer, AutoProcessor, pipeline` ![image](https://github.com/SillyTavern/SillyTavern-Extras/assets/87289390/887dbcdc-efdd-4fc8-bd75-b7d8e95b7e20)

bsakaguchi closed this issue

2026-10-15 18:04:18 +00:00

bsakaguchi commented

2026-10-15 18:04:19 +00:00

Author

Owner

@Cohee1207 commented on GitHub (Sep 9, 2023):

This may be related: https://github.com/coqui-ai/TTS/issues/2455#issuecomment-1565839898
Try reinstalling torch using index-url https://download.pytorch.org/whl/cpu

@Cohee1207 commented on GitHub (Sep 9, 2023): This may be related: https://github.com/coqui-ai/TTS/issues/2455#issuecomment-1565839898 Try reinstalling torch using index-url https://download.pytorch.org/whl/cpu

bsakaguchi commented

2026-10-15 18:04:19 +00:00

Author

Owner

@NyakoFox commented on GitHub (Sep 9, 2023):

Works perfectly, thanks!

@NyakoFox commented on GitHub (Sep 9, 2023): Works perfectly, thanks!

bsakaguchi commented

2026-10-15 18:04:19 +00:00

Author

Owner

@NyakoFox commented on GitHub (Sep 9, 2023):

This is unrelated but I don't think I should open a new issue for another question -- what're the ram requirements? The script gets OOM killed, even if the only enabled module is caption.

@NyakoFox commented on GitHub (Sep 9, 2023): This is unrelated but I don't think I should open a new issue for another question -- what're the ram requirements? The script gets OOM killed, even if the only enabled module is `caption`.

bsakaguchi commented

2026-10-15 18:04:19 +00:00

Author

Owner

@Cohee1207 commented on GitHub (Sep 9, 2023):

Not exactly sure nowadays, but up to 4-6 GB, depending on the chosen modules. Running full unquantized models on CPU unfortunately has high RAM requirements.

If you're just interested in text classification (e.g. emotion sprites), I'm currently working on running the quantized model natively in JavaScript, Summary didn't show itself to be well-performing, but Caption would not be available to be ported, since right now BLIP models can't be run in JS.

@Cohee1207 commented on GitHub (Sep 9, 2023): Not exactly sure nowadays, but up to 4-6 GB, depending on the chosen modules. Running full unquantized models on CPU unfortunately has high RAM requirements. If you're just interested in text classification (e.g. emotion sprites), I'm currently working on running the quantized model natively in JavaScript, Summary didn't show itself to be well-performing, but Caption would not be available to be ported, since right now BLIP models can't be run in JS.

bsakaguchi commented

2026-10-15 18:04:19 +00:00

Author

Owner

@NyakoFox commented on GitHub (Sep 9, 2023):

I was hoping that Caption alone would be small enough to fit within my available RAM (seems to just be 3.8GB available however which is lower than I thought) but it seems like it's not. Thanks anyway!

@NyakoFox commented on GitHub (Sep 9, 2023): I was hoping that Caption alone would be small enough to fit within my available RAM (seems to just be 3.8GB available however which is lower than I thought) but it seems like it's not. Thanks anyway!