[enhancement]: Failsafe update

[jvbrt4]

Is there an existing issue for this?

• I have searched the existing issues

Contact Details

No response

What should this feature add?

Eversince v.5.7 or so I need to manually install Invoke on Windows instead of relying on the Community UI,
Please implement the following command to make sure there are no issues with CUDA installatin:

\resources\bin\uv.exe pip install --python 3.11 --python-preference only-managed invokeai==6.1.0 --force-reinstall --compile-bytecode --index=https://download.pytorch.org/whl/cu128 --prerelease=allow --index-strategy unsafe-best-match. Install NVIDIA driver 570+ and CUDA 12.8

Alternatives

No response

Additional Content

No response

[psychedelicious]

Make sure you are running the latest launcher. Download from https://invoke-ai.github.io/InvokeAI/installation/quick_start/#step-2-download

[jvbrt4]

The latest launcher still fails:

Because only torch<=2.6.0+cu124 is available and invokeai==6.2.0 depends
on torch>=2.7.0,<2.8.dev0, we can conclude that invokeai==6.2.0 cannot
be used.
And because you require invokeai==6.2.0, we can conclude that your
requirements are unsatisfiable.
hint: torch was requested with a pre-release marker (e.g.,
torch>=2.7.0,<2.8.dev0), but pre-releases weren't enabled (try:
--prerelease=allow)
hint: torch was found on https://download.pytorch.org/whl/cu124,
but not at the requested version (torch>=2.7.0,<2.8.dev0). A
compatible version may be available on a subsequent index (e.g.,
https://pypi.org/simple). By default, uv will only consider versions
that are published on the first index that contains a given package, to
avoid dependency confusion attacks. If all indexes are equally trusted,
use --index-strategy unsafe-best-match to consider all versions from
all indexes, regardless of the order in which they were defined.

[psychedelicious]

Did you try repair mode? Checkbox on the last step of the install flow