Strong vs weak LLM routing
A common use case is to route between a strong model and a weak (cheaper, faster) model. We can achieve this using Not Diamond in the following way:
from notdiamond import NotDiamond
client = NotDiamond()
strong_model='openai/gpt-4o' # Choose your strong model
weak_model='openai/gpt-4o-mini' # Choose your weak model
result, session_id, provider = client.chat.completions.create(
messages=[
{"role": "system", "content": "You are a helpful assistant."},
{"role": "user", "content": "Concisely explain merge sort."}
],
model=[strong_model, weak_model],
tradeoff="cost" # Send queries to the weak model when doing so doesn't degrade quality
)
print("LLM called: ", provider.model) # The LLM routed to
print("LLM output: ", result.content) # The LLM response
Updated 3 months ago