Use Orq.ai as the model control layer for OpenAI Agents. Route Agents SDK calls through one OpenAI‑compatible endpoint, capture traces, monitor cost, and manage fallback behavior without rebuilding your agent graphs, MCP tools, or application code.
What is OpenAI?
OpenAI’s Agents SDK is designed for building multi‑step, tool‑using, and MCP‑integrated agents with built‑in tracing and workflows over GPT‑class models. Orq.ai helps those teams focus on agent behavior while centralizing cross‑provider routing, observability, and cost controls around those agents.
Why use Orq with OpenAI
Trace agent behavior end to end
Trace OpenAI Agents runs end to end, including prompts, tool calls (functions, hosted tools, MCP servers), handoffs, guardrails, and model responses in one place. You get clearer visibility into how each agent decided, which tools it used, and where workflows failed, without stitching together logs across multiple services.
Model flexibility without rewiring agents
Test new models, add alternative providers, or assign different model tiers to different agent workflows from Orq.ai, while keeping your Agents SDK definitions and tool wiring stable. Your agents still “think” they’re calling an OpenAI‑style client; Orq handles which underlying model or provider is called.
Evaluate real production runs
Use real traces like agent runs, model calls, tool calls, guardrail decisions to build datasets, compare prompt or model changes, and move from subjective tuning to measurable quality checks. Failed or slow runs can be turned into eval inputs before you roll out changes.
Control spend and access centrally
Track token usage and spend per agent, team, and workflow so you can see which routes drive cost. Add budgets, rate limits, and approved‑model lists at the Orq platform layer instead of enforcing governance separately inside each OpenAI Agents deployment.docs.orq+1
How the integration works
Step 1
Point Agents SDK traffic at Orq.ai’s router
Configure the OpenAI client used by the Agents SDK to talk to Orq’s OpenAI‑compatible router (https://api.orq.ai/v3/router) with your Orq API key. Your agent code still uses the Agents SDK APIs; Orq sits behind the client as the routing layer.
Step 2
Enable tracing from OpenAI Agents to Orq
Keep built‑in Agents SDK tracing and any LangSmith/OpenTelemetry integrations while exporting LLM spans to Orq’s OTLP endpoint. Agents continue to emit traces for runs, tools, and guardrails, and Orq records LLM interactions, latency, and cost metrics, so you see both agent‑level and model‑level traces.
Step 3
Define routes, fallbacks, and policies in Orq
Create routes for key Agents workflows (for example MCP‑backed copilots, multi‑tool agents, or long‑running workflows) and assign them model tiers, fallback chains, and region/data policies. Agents send LLM requests to the configured Orq routes via the OpenAI client, and Orq applies routing before invoking GPT‑5‑class, Gemini, Claude, or other models.
Step 4
Monitor, evaluate, and tune
Once connected, use Orq’s dashboards to watch latency, errors, and cost for Agents SDK workloads, and run evals or experiments on their traces. You can adjust routes, models, or prompt configurations centrally where supported, while the Agents SDK continues to handle MCP tools, guardrails, and workflow logic.
Use Cases
Multi‑agent products with real observability
Trace which OpenAI agent, tool (functions, hosted MCP, local MCP), and model contributed to a failed workflow or unexpected outcome, combining the Agents SDK’s tracing with Orq’s router‑level LLM traces.
Cost‑aware internal tools
Route routine steps like simple QA or routing decisions to lower‑cost models while keeping complex reasoning, multi‑tool workflows, or safety‑critical flows on your strongest routes configured in Orq.
Eval‑driven agent improvements
Reuse failed conversations, tool errors, and guardrail interventions as eval inputs before shipping prompt, tool, or model changes, turning production telemetry into a continuous improvement loop.
Safer experimentation across providers
Test a new model or provider on a small share of Agents SDK traffic, then promote or roll back based on traces and evals, instead of editing each agent or client configuration by hand.
FAQ
Do I have to change my OpenAI code to use Orq.ai?
In many cases, you can repoint the OpenAI client your agents use (base URL, API key, and model name) to Orq’s OpenAI‑compatible router and enable OTEL export to Orq, keeping agent definitions, tools, and MCP wiring the same.
Does Orq.ai replace OpenAI’s own runtime or framework?
No. The Agents SDK remains your framework for building and running agents, tools, MCP servers, and guardrails. Orq.ai sits alongside it as the control plane for models, multi‑provider routing, cost tracking, and additional evaluation across providers. You still design and run agents with OpenAI’s Agents SDK at the core, but you use Orq to see what they cost, which models they should call, and how LLM behavior compares across providers under the same agent workflow.
Can I keep using my existing LLM providers with OpenAI if I move to Orq.ai?
Yes. You can bring your existing OpenAI API key into Orq.ai and route Agents SDK traffic through OpenAI, alongside any new models you add later. That way you centralize access, routing, and tracking without losing the OpenAI setup you already rely on. Bring production controls to OpenAI Agents
