LLM Settings

1
import os
2
from smallestai.atoms.agent.clients.openai import OpenAIClient
3
4
self.llm = OpenAIClient(
5
    model="gpt-4o-mini",
6
    temperature=0.7,
7
    api_key=os.getenv("OPENAI_API_KEY")
8
)

1
# Consistent support agent
2
self.llm = OpenAIClient(model="gpt-4o-mini", temperature=0.3)
3
4
# Engaging sales agent
5
self.llm = OpenAIClient(model="gpt-4o-mini", temperature=0.8)

1
import os
2
from smallestai.atoms.agent.clients.openai import OpenAIClient
3
4
# Example: Using Groq
5
self.llm = OpenAIClient(
6
    model="llama-3.1-70b-versatile",
7
    base_url="https://api.groq.com/openai/v1",
8
    api_key=os.getenv("GROQ_API_KEY")
9
)

1
async def generate_response(self):
2
    response = await self.llm.chat(
3
        messages=self.context.messages,
4
        stream=True
5
    )
6
    
7
    async for chunk in response:
8
        if chunk.content:
9
            yield chunk.content

1
response = await self.llm.chat(
2
    messages=self.context.messages,
3
    stream=True,
4
    tools=self.tool_registry.get_schemas()
5
)

1
from openai import RateLimitError, APIError
2
3
async def generate_response(self):
4
    try:
5
        response = await self.llm.chat(
6
            messages=self.context.messages,
7
            stream=True
8
        )
9
        async for chunk in response:
10
            if chunk.content:
11
                yield chunk.content
12
                
13
    except RateLimitError:
14
        yield "I'm experiencing high demand. Please try again."
15
        
16
    except APIError as e:
17
        logger.error(f"LLM error: {e}")
18
        yield "I'm having trouble. Let me try again."

1
from smallestai.atoms.models import CreateAgentRequest
2
3
agent = CreateAgentRequest(
4
    name="SupportAgent",
5
    synthesizer={
6
        "provider": "smallest",  # Use Waves
7
        "model": "waves_lightning_v2",
8
        "voice_id": "zorin",     # Male, professional
9
        "speed": 1.0
10
    },
11
    # ... other config
12
)