Installation error

[P47Parzival]

I was trying to follow the commands but I guess I have to download deepspeed externally, but it keeps hitting me with this

(llava) PS C:\Users\dhruv\OneDrive\Desktop\Video RAG\LLaVA-NeXT> pip install deepspeed==0.14.4
Collecting deepspeed==0.14.4
Using cached deepspeed-0.14.4.tar.gz (1.3 MB)
Preparing metadata (setup.py) ... error
error: subprocess-exited-with-error

× python setup.py egg_info did not run successfully.
│ exit code: 1
╰─> [34 lines of output]
The system cannot find the file specified.
The system cannot find the file specified.
The system cannot find the file specified.
The system cannot find the file specified.
The system cannot find the file specified.
The system cannot find the file specified.
fatal: not a git repository (or any of the parent directories): .git
Traceback (most recent call last):
File "", line 2, in
File "", line 35, in
File "C:\Users\dhruv\AppData\Local\Temp\pip-install-lu3q8iil\deepspeed_c2804958294d4607922365b148f2f1f3\setup.py", line 212, in
shutil.copytree('.\csrc', '.\deepspeed\ops')
File "C:\Users\dhruv\anaconda3\envs\llava\lib\shutil.py", line 559, in copytree
return _copytree(entries=entries, src=src, dst=dst, symlinks=symlinks,
File "C:\Users\dhruv\anaconda3\envs\llava\lib\shutil.py", line 457, in _copytree
os.makedirs(dst, exist_ok=dirs_exist_ok)
File "C:\Users\dhruv\anaconda3\envs\llava\lib\os.py", line 225, in makedirs
mkdir(name, mode)
FileExistsError: [WinError 183] Cannot create a file when that file already exists: '.\deepspeed\ops'
[WARNING] Torch did not find cuda available, if cross-compiling or running with cpu only you can ignore this message. Adding compute capability for Pascal, Volta, and Turing (compute capabilities 6.0, 6.1, 6.2)
DS_BUILD_OPS=1
[WARNING] deepspeed_not_implemented requires the 'lscpu' command, but it does not exist!
[WARNING] deepspeed_not_implemented attempted to query 'lscpu' after failing to use py-cpuinfo to detect the CPU architecture. 'lscpu' does not appear to exist on your system, will fall back to use -march=native and non-vectorized execution.
[WARNING] deepspeed_not_implemented requires the 'lscpu' command, but it does not exist!
[WARNING] deepspeed_not_implemented attempted to query 'lscpu' after failing to use py-cpuinfo to detect the CPU architecture. 'lscpu' does not appear to exist on your system, will fall back to use -march=native and non-vectorized execution.
[WARNING] cpu_adam requires the 'lscpu' command, but it does not exist!
[WARNING] cpu_adam attempted to query 'lscpu' after failing to use py-cpuinfo to detect the CPU architecture. 'lscpu' does not appear to exist on your system, will fall back to use -march=native and non-vectorized execution.
[WARNING] cpu_adam requires the 'lscpu' command, but it does not exist!
[WARNING] cpu_adam attempted to query 'lscpu' after failing to use py-cpuinfo to detect the CPU architecture. 'lscpu' does not appear to exist on your system, will fall back to use -march=native and non-vectorized execution.
[WARNING] fused_adam requires the 'lscpu' command, but it does not exist!
[WARNING] fused_adam attempted to query 'lscpu' after failing to use py-cpuinfo to detect the CPU architecture. 'lscpu' does not appear to exist on your system, will fall back to use -march=native and non-vectorized execution.
[WARNING] fused_adam requires the 'lscpu' command, but it does not exist!
[WARNING] fused_adam attempted to query 'lscpu' after failing to use py-cpuinfo to detect the CPU architecture. 'lscpu' does not appear to exist on your system, will fall back to use -march=native and non-vectorized execution.
Install Ops={'deepspeed_not_implemented': 1, 'deepspeed_ccl_comm': 1, 'deepspeed_shm_comm': 1, 'cpu_adam': 1, 'fused_adam': 1}
[end of output]

note: This error originates from a subprocess, and is likely not a problem with pip.
error: metadata-generation-failed

× Encountered error while generating package metadata.
╰─> See above for output.

note: This is an issue with the package mentioned above, not pip.
hint: See above for details.
(llava) PS C:\Users\dhruv\OneDrive\Desktop\Video RAG\LLaVA-NeXT> pip uninstall deepspeed -y
WARNING: Skipping deepspeed as it is not installed.

[onestardao]

This looks like No.16 — Pre-Deploy Collapse (Safety Boundary) from our 16-item AI failure list.
Your log shows a classic toolchain/version mismatch before first run: building deepspeed from source on Windows (egg_info fails, “not a git repo”, ispc/compiler missing, CUDA/Torch version hints). The app never reaches runtime; it collapses at install.

Quick fixes that usually unblock this pattern:

• Use WSL2/Ubuntu or Linux and install prebuilt wheels (avoid building DeepSpeed on Windows).
• Pin a compatible trio: Python ↔ Torch+CUDA ↔ DeepSpeed (install Torch with the exact CUDA build first, then pip install deepspeed==<compatible>).
• If DS isn’t essential, disable it temporarily and run CPU or plain Torch to verify the rest of the stack.
• Lock versions in requirements.txt and add a preflight script to assert compiler/CUDA presence before install.

We’ve helped 80+ developers resolve this class of issue with Safety-Boundary checks.
MIT-licensed project, 600+ GitHub stars in ~60 days.
If you’re interested, reply here and I can share the link.