Integrations
Fits the stack your team already ships
Import one SDK for complete evaluation across any LLM provider, agent framework, or orchestration layer.
LLM Providers
Evaluate every model fully.
Add three code lines to instrument any provider call. The evaluator runs after each completion.
OpenAI
GPT-4o, GPT-4 Turbo, GPT-3.5
Anthropic
Claude 3.5, Claude 3, Claude Instant
Google AI
Gemini 1.5 Pro, Gemini Flash
Mistral
Mistral Large, Mixtral
Cohere
Command R+, Command
Bedrock
Claude via AWS, Llama 3 via AWS
Agent Frameworks
Built-in multi-agent orchestration support
Traces every hop and scores the full chain, not only the final answer.
LangChain
Callbacks, Agents, LCEL
LlamaIndex
RAG pipelines, agents
AutoGen
Multi-agent conversations
CrewAI
Crew tasks, tool use
LangGraph
State graphs, branching flows
Haystack
Pipelines, document stores
Quick install
Full observability in three lines
Install the SDK, wrap your client, set your rubric. That completes integration.
import futureagi
from openai import OpenAI
# Wrap your existing client - zero other changes
client = futureagi.wrap(
OpenAI(),
api_key="fagi_...",
rubric="Responses must be factually accurate and not hallucinate."
)
# Your existing call - now evaluated on every request
response = client.chat.completions.create(
model="gpt-4o",
messages=[{"role": "user", "content": user_message}]
)
Missing an integration? Tell us.
We add integrations every two weeks. Contact us, and it may ship next sprint.