[PR-1090] Agentic Class

[luv-bansal]

Pull request overview

This PR introduces a new AgenticModelClass that extends OpenAIModelClass to enable agentic behavior by integrating LLMs with MCP (Model Context Protocol) servers. The class provides tool discovery, execution, and iterative tool calling capabilities for both chat completions and responses endpoints, supporting both streaming and non-streaming modes.

LLM with MCP servers client example

from openai import OpenAI
import os
client = OpenAI(
    base_url="https://api.clarifai.com/v2/ext/openai/v1",
    api_key= os.environ['CLARIFAI_PAT']
)
mcp_servers = [
    "https://api.clarifai.com/v2/ext/mcp/v1/users/clarifai/apps/mcp/models/calculator-mcp-server",
    "https://api.clarifai.com/v2/ext/mcp/v1/users/clarifai/apps/mcp/models/web-search-mcp-server",
]
completion = client.chat.completions.create(
    model="https://clarifai.com/clarifai/agentic-model/models/gpt-5_1",
    messages=[{"role": "user", "content": "What's the weather in Los Angeles, California?"}],
    extra_body={"mcp_servers": mcp_servers},
    max_completion_tokens=10000,
    stream=True
)
for chunk in completion:
    if chunk.choices and len(chunk.choices) > 0 and hasattr(chunk.choices[0].delta, 'content') and chunk.choices[0].delta.content:
        print(chunk.choices[0].delta.content, end="", flush=True)
    elif chunk.choices and len(chunk.choices) > 0 and hasattr(chunk.choices[0].delta, 'reasoning_content') and chunk.choices[0].delta.reasoning_content:
        print(chunk.choices[0].delta.reasoning_content, end="", flush=True)

[Copilot]

Pull request overview

This PR introduces a new AgenticModelClass that extends OpenAIModelClass to enable agentic behavior by integrating LLMs with MCP (Model Context Protocol) servers. The class provides tool discovery, execution, and iterative tool calling capabilities for both chat completions and responses endpoints, supporting both streaming and non-streaming modes.

Key Changes

• Adds MCP server connection and tool discovery with retry logic and resource cleanup
• Implements token usage accumulation across multiple API calls in agentic flows
• Provides streaming and non-streaming support for iterative tool execution with both chat completions and responses API endpoints

[Copilot]

Pull request overview

Copilot reviewed 1 out of 1 changed files in this pull request and generated 16 comments.

[Copilot]

Pull request overview

Copilot reviewed 1 out of 1 changed files in this pull request and generated 33 comments.

[Copilot]

@luv-bansal I've opened a new pull request, #875, to work on those changes. Once the pull request is ready, I'll request review from you.

[Copilot]

@luv-bansal I've opened a new pull request, #876, to work on those changes. Once the pull request is ready, I'll request review from you.

[Copilot]

Pull request overview

Copilot reviewed 2 out of 2 changed files in this pull request and generated 15 comments.

[github-actions]

Package	Line Rate	Health
clarifai	45%	❌
clarifai.cli	58%	➖
clarifai.cli.templates	46%	❌
clarifai.client	67%	➖
clarifai.client.auth	67%	➖
clarifai.constants	100%	✔
clarifai.datasets	100%	✔
clarifai.datasets.export	80%	✔
clarifai.datasets.upload	75%	➖
clarifai.datasets.upload.loaders	37%	❌
clarifai.models	100%	✔
clarifai.modules	0%	❌
clarifai.rag	0%	❌
clarifai.runners	54%	➖
clarifai.runners.models	65%	➖
clarifai.runners.pipeline_steps	41%	❌
clarifai.runners.pipelines	77%	➖
clarifai.runners.utils	62%	➖
clarifai.runners.utils.data_types	72%	➖
clarifai.schema	100%	✔
clarifai.urls	60%	➖
clarifai.utils	60%	➖
clarifai.utils.evaluation	67%	➖
clarifai.workflows	95%	✔
Summary	63% (9141 / 14541)	➖

Minimum allowed line rate is 50%