Models · Audio
GPT Audio
by OpenAI
The gpt-audio model is OpenAI's first generally available audio model.
Our verdict
We have not reviewed GPT Audio yet
No one at Enterprise DNA has checked this listing against its sources. The description comes from public listings gathered automatically. Any fact below shows where it came from and when it was last checked.
From the listing
About GPT Audio
The gpt-audio model is OpenAI's first generally available audio model. The new snapshot features an upgraded decoder for more natural sounding voices and maintains better voice consistency.
Best for
Audio input and spoken responses
Notes
GPT Audio is listed in the EDNA Models directory from the OpenRouter catalogue.
Pros
- Prototyping a spoken assistant with explicit turn boundaries
- text, audio input; text, audio output on the listed OpenRouter route
Cons
- Assuming batch audio implies realtime transport support
- Estimating audio requests with only text-token prices
- Base text-token rates exclude other modalities, tools, caching, retries and any hosting-specific charges. Check actual usage before comparing cost per successful task.
Key facts
What we have checked
- Input price
- $2.50 per million tokens
- Output price
- $10.00 per million tokens
- Price on openai
- $2.50 per million input tokens
- Price on openai
- $10.00 per million output tokens
- Latest version
- openai/gpt-audio
- Time to first token (ms) on openai
- 627
- Output speed (tokens a second) on openai
- 71
- Tokens served on OpenRouter in a day
- 14.8M (for 10 Nov 2026)
- Tool calls on OpenRouter in a day
- 122 (for 10 Nov 2026)
Each value shows when we last confirmed it and where it came from. Values we have not confirmed are left out, never guessed.
API access
How to call it
Call via OpenRouter with model id `openai/gpt-audio`.
- openrouter ·
openai/gpt-audio· Docs
OpenRouter id: openai/gpt-audio
Price
What GPT Audio costs, over time
$2.50 to $2.50
$10.00 to $10.00
Prices by host
| Host | Price | Over time | Checked |
|---|---|---|---|
| openaiInput | $2.50 per million input tokensSince 24 Oct 2026 | No change recorded | OpenRouter models API |
| openaiOutput | $10.00 per million output tokensSince 24 Oct 2026 | No change recorded | OpenRouter models API |
Usage
How much GPT Audio is used
6.6M to 14.8M
Sources
Where this comes from
- OpenRouter models APIRegistry
- OpenRouter models APIThird party
- OpenRouter models APIOfficial
A source shows what was published when we checked it. It does not certify security, performance or a working installation. Read our method.
Changelog
What changed
No changes recorded yet. Releases, price changes and deprecations appear here, with their dates, once we track them for this record.
USAGE / OPENROUTER
How much GPT Audio is used
Tokens processed through OpenRouter each day. It shows what developers actually run, not every company's private traffic.
- Tokens, last 7 days
- 86.5M ▼21.1%
- Requests, last 7 days
- 145K
- Tool calls, last 7 days
- 1.2K
- Price per million tokens
- $2.5 in · $10 out
MODEL FIELD GUIDE / SOURCE EVIDENCE
What you can do with GPT Audio.
Specifications reviewed; no EDNA task evaluation has been run. Sources checked 2026-11-07. Hosting routes and app features can differ.
openai/gpt-audio- Tool calling
- Listed support
- Structured output
- Listed support
- Reasoning controls
- Not established
- Availability
- available
- Price basis
- token
- Licence verification
- Check the developer licence; the model name is insufficient
- Context length
- 128K tokens
- Knowledge cutoff
- Not established
- Developer release date
- Not established
- First seen in the catalogue
- 19 Jan 2026
- Listed retirement date
- Not established
Base text-token rates exclude other modalities, tools, caching, retries and any hosting-specific charges. Check actual usage before comparing cost per successful task.
Consider it for
- Prototyping a spoken assistant with explicit turn boundaries
- Testing text and audio response consistency
Watch for
- Assuming batch audio implies realtime transport support
- Estimating audio requests with only text-token prices
PRACTICAL EXERCISE / NOT RUN BY ENTERPRISE DNA
Try the model on a small, checkable task.
Read aloud: Your order contains three items and arrives on Tuesday. Use a calm neutral voice and add no other information.
Check the result: Listen for the exact item count and delivery day; compare any text response to the speech.
Compare another approach
Inspect the sources and their scope
Developer documentation for model family or task. OpenRouter route-specific properties remain catalogue claims; EDNA has not executed a task evaluation.
Checked 2026-10-19openai/gpt-audio: listed route capabilities, availability and prices; provider endpoints can differ.
Checked 2026-11-07Source for first seen, tool calling, context length, input, output, structured output.
Checked 2026-11-07FROM SPECIFICATIONS TO YOUR WORK
Is this the right model for your job?
Start with a realistic sample. These workflows match the model’s listed capabilities, not a verified result on your task.
What would it cost?
Pick a common job and how many times a month you run it. Token sizes are our assumption for a typical run, shown under the result.
Enter a whole number of runs from 1 to 100,000,000.
Working out the cost...
Token prices only. Per-request, image, tool, storage and subscription fees are not included. One token is about three quarters of an English word. Real counts vary by model and language. How tokens are counted ↗
Check the source
Price date: 10 Nov 2026, from OpenRouter models API. Prices are USD API token rates. Your app subscription, uploads, browsing and tools are separate.
Model source & availability ↗Understand the evidence
Listed capabilities and third-party benchmark scores do not establish Enterprise DNA task-test results. Validate accuracy and total cost on your own sample.
Get the free Developer’s Field Guide
A 27-page field guide to the AI coding workflow with Claude. Claude Code, MCP servers, the prompt patterns that work, and what to delegate. Free.
Enter your work email. We send it straight over, plus a few short notes worth knowing. Unsubscribe any time.