ADDED: multi-LLM support via LLMFactory (fully backward compatible) and OptoPrimeMulti multi_llm option

[doxav]

ADDED: multi-LLM support via LLMFactory (fully backward compatible) and implementation demonstration in OptoPrimeMulti and associated test

[chinganc]

Can you change this PR to merge into experimental branch?

[chinganc]

What is the expected way to call register_profile so as to modify the behavior of LLM (for default llm applications)? Is the following the intended usage?

from opto.utils.llm import LLM, LLMFactory

LLMFactory.register_profile(new_backend_1,  new_param_1)
LLMFactory.register_profile(new_backend_2,  new_param_2)

llm_1 = LLM(profile=new_param_1) 
llm_2 = LLM(profile=new_param_2) 

[doxav]

Correct Usage

# Register new profiles
LLMFactory.register_profile("custom_openai", "LiteLLM", model="gpt-4o", temperature=0.7, max_tokens=2000)
LLMFactory.register_profile("custom_claude", "LiteLLM", model="claude-3-5-sonnet-latest", temperature=0.3)
LLMFactory.register_profile("local_llama", "CustomLLM", model="llama-3.1-70b")

# Use the registered profiles
llm_1 = LLM(profile="custom_openai") 
llm_2 = LLM(profile="custom_claude")
llm_3 = LLM(profile="local_llama")

Pre-defined Profiles Available

The code comes with these built-in profiles:

llm_default = LLM(profile="default")     # gpt-4o-mini
llm_premium = LLM(profile="premium")     # gpt-4  
llm_cheap = LLM(profile="cheap")         # gpt-4o-mini
llm_fast = LLM(profile="fast")           # gpt-3.5-turbo-mini
llm_reasoning = LLM(profile="reasoning") # o1-mini

You can override those built-in profiles:

LLMFactory.register_profile("default", "LiteLLM", model="gpt-4o", temperature=0.5)
LLMFactory.register_profile("premium", "LiteLLM", model="o1-preview", max_tokens=8000)
LLMFactory.register_profile("cheap", "LiteLLM", model="gpt-3.5-turbo", temperature=0.9)
LLMFactory.register_profile("fast", "LiteLLM", model="gpt-3.5-turbo", max_tokens=500)
LLMFactory.register_profile("reasoning", "LiteLLM", model="o1-preview")

Examples with Different Backends

# Register custom profiles for different use cases
LLMFactory.register_profile("advanced_reasoning", "LiteLLM", model="o1-preview", max_tokens=4000)
LLMFactory.register_profile("claude_sonnet", "LiteLLM", model="claude-3-5-sonnet-latest", temperature=0.3)
LLMFactory.register_profile("custom_server", "CustomLLM", model="llama-3.1-8b")

# Use in different contexts
reasoning_llm = LLM(profile="advanced_reasoning")  # For complex reasoning
claude_llm = LLM(profile="claude_sonnet")          # For Claude responses
local_llm = LLM(profile="custom_server")           # For local deployment

# Single LLM optimizer with custom profile
optimizer1 = OptoPrime(parameters, llm=LLM(profile="advanced_reasoning"))

# Multi-LLM optimizer with multiple profiles
optimizer2 = OptoPrimeMulti(parameters, llm_profiles=["cheap", "premium", "claude_sonnet"], generation_technique="multi_llm")