NeMo LLM Python Client

0.3.5 verified Fri May 01 auth: no python

Python client library for NVIDIA NeMo LLM API. Version 0.3.5. The library enables interaction with NeMo LLM services for text generation. Release cadence is irregular.

pip install nemollm

Common errors

error ImportError: cannot import name 'NeMoLLM' from 'nemollm' ↓

cause The library version may not support the expected class name. Ensure correct spelling and version.

fix

Verify installation: pip show nemollm. Import as: from nemollm import NeMoLLM

error AttributeError: 'NeMoLLM' object has no attribute 'generate' ↓

cause API method may be named differently, e.g., 'complete' or 'chat'.

fix

Check the documentation for the correct method name. In older versions it might be 'complete'.

Warnings

gotcha The NeMoLLM client does not automatically retry on rate limits. You may need to implement retry logic with backoff. ↓

fix Wrap calls with retry logic using tenacity or similar.

gotcha The API key must be passed as a keyword argument, not positional. Omitting the parameter name may cause unexpected error. ↓

fix Always use client = NeMoLLM(api_key='your_key')

Imports

NeMoLLM
```
from nemollm import NeMoLLM
```
No common wrong import pattern.

Quickstart

Initialize the client with an API key and generate text.

from nemollm import NeMoLLM

api_key = os.environ.get('NEMO_API_KEY', '')
client = NeMoLLM(api_key=api_key)
response = client.generate("Hello, how are you?")
print(response)