Apply your own routing policies, reduce token costs automatically, and see every routing decision in real time with Merge Gateway.
What GPT Audio costs to run
Test GPT Audio
with Gateway’s Simulator
See a prompt's output, token spend, latency, and more with GPT Audio.
Route requests to GPT Audio in minutes
1$ pip install merge-gateway-sdk1from merge_gateway import MergeGateway
2
3client = MergeGateway(api_key="YOUR_API_KEY")
4
5response = client.responses.create(
6 model="openai/gpt-5.2",
7 input=[
8 {"type": "message", "role": "system", "content": "You are a helpful programming tutor. Explain the concepts clearly with practical examples."},
9 {"type": "message", "role": "user", "content": "Explain the concept of recursion in programming with a simple set of examples."},
10 ],
11)
12
13print(response.output[0].content[0].text)1response = client.responses.create(
2 model="anthropic/claude-sonnet-4-20250514",
3 input=[
4 {"type": "message", "role": "system", "content": "You are a helpful programming tutor. Explain the concepts clearly with practical examples."},
5 {"type": "message", "role": "user", "content": "Explain the concept of recursion in programming with a simple set of examples."},
6 ],
7)1from openai import OpenAI
2
3client = OpenAI(
4 api_key="YOUR_API_KEY",
5 base_url="https://api-gateway.merge.dev/v1/openai",
6)1response = client.chat.completions.create(
2 model="gpt-5.2",
3 messages=[
4 {"role": "system", "content": "You are a helpful programming tutor. Explain the concepts clearly with practical examples."},
5 {"role": "user", "content": "Explain the concept of recursion in programming with a simple set of examples."},
6 ],
7)
8
9print(response.choices[0].message.content)1npm install merge-gateway-ai-sdk-provider ai1import { createMergeGateway } from "merge-gateway-ai-sdk-provider";
2
3const gateway = createMergeGateway({
4 apiKey: "YOUR_API_KEY",
5});1import { generateText } from "ai";
2
3const { text } = await generateText({
4 model: gateway("openai/gpt-4o"),
5 prompt: "Explain the concept of recursion in programming with a simple set of examples.",
6});
7
8console.log(text);1import { createOpenAI } from "@ai-sdk/openai";
2
3const gateway = createOpenAI({
4 apiKey: "YOUR_API_KEY",
5 baseURL: "https://api-gateway.merge.dev/v1/ai-sdk",
6});
7
8// All generateText/streamText calls work unchanged1from anthropic import Anthropic
2
3client = Anthropic(
4 api_key="YOUR_API_KEY",
5 base_url="https://api-gateway.merge.dev/v1/anthropic",
6)
7
8message = client.messages.create(
9 model="claude-sonnet-4-20250514",
10 max_tokens=1024,
11 messages=[
12 {"role": "user", "content": "Explain the concept of recursion in programming with a simple set of examples."},
13 ],
14)
15
16print(message.content[0].text)Explore other models available in Merge Gateway
GPT Audio FAQ
Heading
What other models does OpenAI offer?
OpenAI ships general-purpose text flagships, reasoning models, and modality-specific models for audio and images. Here are some other models OpenAI supports:
- GPT-5.5: the current general-purpose flagship, strong across reasoning, coding, and agentic tasks
- GPT-5: the prior general-purpose flagship, still widely used across production text workloads
- GPT Audio Mini: a smaller, cheaper audio model for high-volume speech tasks where cost matters more than peak quality
How does GPT Audio differ from OpenAI's other models?
GPT Audio is OpenAI's speech-native model, built to take audio in and produce audio out rather than to reason over text.
- Modality: accepts spoken audio and returns speech, and it handles transcription and audio understanding, unlike the text-first GPT-5 line
- Use case fit: aimed at voice applications like assistants, call handling, and audio summarization, not text-heavy reasoning
- Quality vs GPT Audio Mini: higher audio quality and comprehension at a higher price than the Mini tier
- Turn-based vs streaming: processes discrete audio requests, where GPT Realtime is tuned for live, low-latency streaming conversations
Reach for GPT Audio when your product needs quality speech understanding and generation over the API without building a separate speech-to-text and text-to-speech pipeline.
What models should I consider using alongside GPT Audio?
No single model is optimal for every task. Here are models worth pairing with GPT Audio depending on what your product needs:
- GPT-5.5 for the text reasoning behind a voice agent once speech has been turned into text
- Gemini 3.5 Flash for high-volume, low-cost multimodal steps that mix audio and text
- Claude Opus 4.8 for complex reasoning or drafting over transcribed content
- Gemini 3.1 Pro for long-context analysis of long recordings or multi-speaker transcripts
What are the challenges of using GPT Audio in my product?
Like any production LLM, GPT Audio comes with tradeoffs worth planning for:
- Audio token cost: audio tokens are priced higher than text, so high-volume voice traffic compounds quickly without active budgeting
- Latency: turn-based audio processing adds round-trip latency, so plan the UX around it or use a streaming model for live conversation
- Modality scope: it's built for speech, so pair it with a text model for heavy reasoning rather than pushing everything through audio
- Voice data sensitivity: spoken audio can carry personal or biometric data, which raises retention, consent, and DLP requirements
- Provider dependency: relying only on OpenAI is fragile during outages or model retirements
Why should I use Merge Gateway to route LLM requests with GPT Audio and every other model?
Routing GPT Audio through Merge Gateway pairs the model with the control layer around it:
- One API, every provider: Call GPT Audio and every other major model from a single endpoint and key, switching models with a string change rather than a code change
- Intelligent routing and automatic failover: Merge fails over around OpenAI outages on its own, and routing on cost, latency, or quality can lower spend by 40 to 60% with no code edits
- Cost governance: Cap audio spend with project budgets, and see every request attributed by model, project, and tag in one cross-provider dashboard
- Build Your Own Router: Define "best" for your traffic with curated benchmarks or your own evals, and let the router choose per request and explain why
- Security and compliance controls: Run DLP and prompt injection checks before requests reach OpenAI, and enforce per-project model and region rules centrally
How can I start routing requests to GPT Audio via Merge Gateway?
Getting GPT Audio running through Merge Gateway takes a few minutes:
1. Create an account and get your API key from the dashboard.
2. Install the Merge Gateway SDK: run pip install merge-gateway-sdk (Python) or npm install merge-gateway-sdk (Node). Alternatively, if you're already using the OpenAI SDK, set base_url = "https://api-gateway.merge.dev/v1/openai" and your existing code works as-is.
3. Make your first request using the provider/model format. For GPT Audio, the model string is openai/gpt-audio (confirm the exact slug in the dashboard). Swap the model string to route to any other provider without changing anything else.
4. Configure a routing policy in the dashboard to set failover behavior, cost limits, and optimization strategy. Your first policy can be as simple as naming GPT Audio as primary with one fallback.
Full setup instructions and SDK references are in the Merge Gateway docs.
Try GPT Audio through Merge Gateway
Route, observe, and control AI requests across providers from one API.
