Models API

Agent-grade model providers live in elsai-model (elsai Core). Pass a *Model instance directly to Agent(model=...). The agent loop calls the model on every turn, handling tool specs, system prompts, and streaming events uniformly across providers.

For standalone invoke / stream usage without agents, see LLM Models. For provider setup and credentials, see Model Providers.

Installation

pip install --extra-index-url https://elsai-agents.elsai.ai/root/ elsai-agents==0.3.0
pip install --extra-index-url https://core-packages.elsai.ai/root/elsai-model/ elsai-model==2.0.0

Some providers need optional extras on elsai-model — see LLM Models — Optional package extras.

---

Usage with Agent

from elsai import Agent
from elsai_model.openai import OpenAIModel

model = OpenAIModel(
    model_id="gpt-4o",
    client_args={"api_key": "..."},  # or OPENAI_API_KEY env var
    params={"temperature": 0.7},
)

agent = Agent(model=model, system_prompt="You are a concise assistant.")
result = agent("What is 17 + 28?")

---

Supported providers

Provider	Model class	Import module
OpenAI	OpenAIModel	elsai_model.openai
Azure OpenAI	AzureOpenAIModel	elsai_model.azure_openai
Amazon Bedrock	BedrockModel	elsai_model.bedrock
Anthropic (direct)	AnthropicModel	elsai_model.anthropic
Anthropic via Bedrock SDK	AnthropicBedrockModel	elsai_model.anthropic_bedrock
Google Gemini	GeminiModel	elsai_model.gemini
LiteLLM	LiteLLMModel	elsai_model.litellm
Ollama	OllamaModel	elsai_model.ollama
Mistral	MistralModel	elsai_model.mistral
Writer	WriterModel	elsai_model.writer
Meta Llama API	LlamaAPIModel	elsai_model.llamaapi
llama.cpp server	LlamaCppModel	elsai_model.llama_cpp
OpenAI Responses	OpenAIResponsesModel	elsai_model.openai_responses
SageMaker	SageMakerAIModel	elsai_model.sagemaker

Per-provider constructor parameters, environment variables, and examples are documented in LLM Models and the Model Providers concept pages.

---

Common parameters

All agent-grade *Model classes share this constructor shape:

Parameter	Description
model_id	Provider model or deployment ID
client_args	SDK client options (API keys, endpoints, regions)
params	Generation settings (temperature, max_tokens, …)

---

BedrockModel (example)

from elsai_model.bedrock import BedrockModel

BedrockModel(
    model_id="us.anthropic.claude-3-5-sonnet-20241022-v2:0",
    region_name="us-east-1",
    max_tokens=4096,
    temperature=0.3,
)

See Amazon Bedrock and LLM Models — Amazon Bedrock.

---

OpenAIModel (example)

from elsai_model.openai import OpenAIModel

OpenAIModel(
    model_id="gpt-4o",
    client_args={"api_key": "..."},
    params={"temperature": 0.7},
)

See OpenAI and LLM Models — OpenAI.

---

Legacy connectors

Legacy v1.4.x *Connector classes and GeminiService are not supported with Agent. Pass agent-grade *Model classes from elsai-model==2.0.0 instead. See elsai Core v3.0 release notes.

---

Related

• Model Providers — install, credentials, and provider selection
• LLM Models — full provider API reference for elsai-model
• Agent API — Agent(model=…) and runtime options