#401·llama3

Groq API error

Author: PhoenixAlpha23Created Jun 15, 2025Updated Jun 15, 2025

Unexpected Behaviour

While trying to access the llama-3.3-70b-versatile model via groq encountered this error.

Minimal reproducible example

python

from langchain_groq import ChatGroq
from langchain.prompts import PromptTemplate
from langchain.chains import LLMChain
load_dotenv()

GROQ_API_KEY = os.getenv("GROQ_API_KEY")
client = Groq(api_key=GROQ_API_KEY)

MODEL_NAME = "llama-3.3-70b-versatile"   

# prompt template
prompt_template = PromptTemplate(
    input_variables=["question"],
    template="""
You are a helpful assistant. Answer the following question clearly and concisely:

Question: {question}

Answer:"""
)

# Initialize the LLaMA model via Groq
llm = ChatGroq(
    api_key=GROQ_API_KEY,
    model=MODEL_NAME,
    temperature=0.2,
    max_tokens=2048
)

# Wrap it into an LLMChain with the prompt
llm_chain = LLMChain(
    llm=llm,
    prompt=prompt_template,
    verbose=True  
)

# Example usage
user_question = "What is the capital of Maharashtra?"
response = llm_chain.run(user_question)

print("Response:")
print(response)

Output

Image

Runtime Environment

  • Model: llama-3.3-70b-versatile
  • Using via huggingface?: No
  • Platform: Streamlit

Additional context The llama-3.3-70b-versatile model was being used via groq API (free version). This error occured between 12:00 IST and 12:30 ISt on Friday 13th June 2025. Mistake in outputs was also noticed such as chinese alphabets in between a marathi language answer. A custom PromptTemplate was used while inferencing and was followed accurately until before this error showed up. The quality of results seem to have taken a hit after this.