CUDA error while running "model.fit"

[NekoClaviS]

I am facing some cupy cuda error while running the following code:

CUPY_DUMP_CUDA_SOURCE_ON_ERROR =1
CUPY_DISABLE_JITIFY_CACHE = 1
model.fit(min_iter=30, max_iter=35)
model.plot_convergence()

The above code runs for 1 iteration and then gives the following error:

6 errors detected in the compilation of "cupy_jitify_exercise".

RuntimeError                              Traceback (most recent call last)
Cell In[80], line 3
      1 CUPY_DUMP_CUDA_SOURCE_ON_ERROR =1
      2 CUPY_DISABLE_JITIFY_CACHE = 1
----> 3 model.fit(min_iter=30, max_iter=35)
      4 model.plot_convergence()

File ~/anaconda3/envs/scenvi/lib/python3.9/site-packages/SEACells/core.py:587, in SEACells.fit(self, max_iter, min_iter, initial_archetypes)
    585 if max_iter < min_iter:
    586     raise ValueError("The maximum number of iterations specified is lower than the minimum number of iterations specified.")
--> 587 self._fit(max_iter=max_iter, min_iter=min_iter, initial_archetypes=initial_archetypes, initial_assignments=None)

File ~/anaconda3/envs/scenvi/lib/python3.9/site-packages/SEACells/core.py:544, in SEACells._fit(self, max_iter, min_iter, initial_archetypes, initial_assignments)
    532 def _fit(self, max_iter: int = 50, min_iter:int=10, initial_archetypes=None, initial_assignments=None):
    533     """
    534     Compute archetypes and loadings given kernel matrix K. Iteratively updates A and B matrices until maximum
    535     number of iterations or convergence has been achieved.
   (...)
    542 
    543     """
--> 544     self.initialize(initial_archetypes=initial_archetypes, initial_assignments=initial_assignments)
    546     converged = False
    547     n_iter = 0
...
File cupy/cuda/jitify.pyx:192, in cupy.cuda.jitify._init_cupy_headers_from_scratch()

File cupy/cuda/jitify.pyx:264, in cupy.cuda.jitify.jitify()

RuntimeError: Runtime compilation failed

I have tried multiple things and have even tried rebuilding my anaconda venv, but unable to fix this, it would be really helpful if someone could help with the same.