[BUG] Windows model downloads not working as expected about gpt4all-ts HOT 3 OPEN

Hello there,
I was looking around at the issue and I discovered that moving the gpt4all-lora-quantized-win64.exe file (renamed to gpt4all.exe) from the gpt4all chat folder to .nomic/ directory appears to have resolved the problem.

from gpt4all-ts.

I can confirm, the root cause is that the source code downloads the .exe file as gpt4all (without the file extension) but the second you try to run an executable in Windows-based CLIs that doesn't explicitly have .exe or other file type causes the child_process that spawns to not know what default behavior to assume with the file and thus it will fail to execute.

Simply by keeping the name "as-is" and running the exe file that you download as an exe file should resolve the problem.
The workaround I used because I didn't have to worry about breaking anyone else's computer was this:

...
// Added .exe here:
this.executablePath = `${os.homedir()}/.nomic/gpt4all.exe`;
...

from gpt4all-ts.

I opened a pr for this: #11

from gpt4all-ts.