---
format: "aidr-story-markdown/v1"
id: "2ea1c1b871e1a8a02b87d69e582761904409d72003d975f13883f99a61c148a0"
canonical_url: "https://aidr.today/2ea1c1b8?lang=en"
title: "OpenAI GPT-Live-1 Tops Speech Index With 81.5 Debut Score"
lang: "en"
requested_lang: "en"
available_langs: ["en","vi"]
translation_fallback: null
fallback_fields: []
published_at: "2026-09-15T06:03:12.000Z"
category: "Models"
topics: ["openai","gpt","text-to-speech","benchmark","inference","api"]
source_urls: ["https://huggingnews.com/ai/openai-gpt-live-1-tops-speech-index-with-815-debut-score-05a375a9","https://x.com/ArtificialAnlys/status/2099698254414029207","https://x.com/OpenAIDevs/status/2099720632481321397","https://x.com/ArtificialAnlys/status/2099698257484292184","https://x.com/ArtificialAnlys/status/2099698259812139317","https://x.com/ArtificialAnlys/status/2099698264618729896","https://marketbrief.now/ai/openai-gpt-live-1-tops-speech-index-with-815-debut-score-05a375a9"]
summary: "The new full duplex speech model from OpenAI delegates reasoning and tool use to a backend text model while streaming audio via API. Using Astra at medium reasoning effort, the GPT-Live-1 model achieved 81.5 on the Artificial Analysis Speech to Speech Index, placing it ahead of Grok Voice Think Fast 2.0 High at 81.3. Costs for the model are $4.47 per hour for the Sol backend and $5.83 per hour for Astra, inclusive of session charges and token usage. Gemini 3.1 Flash Live Minimal leads conversational preference with 1,096 Elo, while GPT-Live-1 Sol ranks third at 1,053. GPT-Live-1 holds the top two spots on the Tau Voice agentic benchmark at 67.9% and 59.3%, though these figures are based on a single trial. Average time to first audio for the model is 1.24 to 1.34 seconds, compared with 0.70 seconds for Grok."
---

# OpenAI GPT\-Live\-1 Tops Speech Index With 81\.5 Debut Score

> [Open the canonical story](<https://aidr.today/2ea1c1b8?lang=en>)

**Published:** 2026-09-15T06:03:12.000Z
**Category:** Models
**Topics:** openai, gpt, text\-to\-speech, benchmark, inference, api

## Summary

The new full duplex speech model from OpenAI delegates reasoning and tool use to a backend text model while streaming audio via API\. Using Astra at medium reasoning effort, the GPT\-Live\-1 model achieved 81\.5 on the Artificial Analysis Speech to Speech Index, placing it ahead of Grok Voice Think Fast 2\.0 High at 81\.3\. Costs for the model are $4\.47 per hour for the Sol backend and $5\.83 per hour for Astra, inclusive of session charges and token usage\. Gemini 3\.1 Flash Live Minimal leads conversational preference with 1,096 Elo, while GPT\-Live\-1 Sol ranks third at 1,053\. GPT\-Live\-1 holds the top two spots on the Tau Voice agentic benchmark at 67\.9% and 59\.3%, though these figures are based on a single trial\. Average time to first audio for the model is 1\.24 to 1\.34 seconds, compared with 0\.70 seconds for Grok\.

## Sources

- [Story source](<https://huggingnews.com/ai/openai-gpt-live-1-tops-speech-index-with-815-debut-score-05a375a9>)
- [Story source](<https://x.com/ArtificialAnlys/status/2099698254414029207>)
- [Supporting source](<https://x.com/OpenAIDevs/status/2099720632481321397>)
- [Supporting source](<https://x.com/ArtificialAnlys/status/2099698257484292184>)
- [Supporting source](<https://x.com/ArtificialAnlys/status/2099698259812139317>)
- [Supporting source](<https://x.com/ArtificialAnlys/status/2099698264618729896>)
- [Story source](<https://marketbrief.now/ai/openai-gpt-live-1-tops-speech-index-with-815-debut-score-05a375a9>)