← Back to Voice models
OpenAINotable
GPT-Realtime-2.1 Mini
Released Jul 6, 2026
A distilled, faster and cheaper version of GPT-Realtime-2.1 for realtime voice apps, with reasoning and tool use.
- Price per minute
- —
- Pricing
- Per 1M tokens: text $0.60 in / $0.06 cached / $2.40 out; audio $10 in / $0.30 cached / $20 out; image $0.80 in / $0.08 cached.
- Inputs
- audio, text, image
- Tool calling
- Yes
- Languages
- —
- Voices
- 10
- Big Bench Audio
- 74.7%
- Access
- Generally available
- Licence
- Proprietary
Strengths & weaknesses
Strengths
- Audio at $10 in / $20 out per 1M tokens, under a third of GPT-Realtime-2.1
- Function calling and reasoning tokens
- OpenAI's named replacement for gpt-realtime-mini and its TTS models
Weaknesses
- Big Bench Audio 74.7% at High and 62.7% at Minimal (AA), far below the full model
Notable: frontier-lab clause. AA lists 'GPT-Realtime-2.1 Mini High' at 74.7% and 'Mini Minimal' at 62.7% (S2S Index 54.4). OpenAI's deprecations page names it as the replacement for gpt-realtime-mini (shutdown 2027-01-20) and for tts-1, tts-1-hd and gpt-4o-mini-tts (shutdown 2027-01-06). 128K context, 32K max output. Voices: the same 10 Realtime voices.
Links
API model strings
- openai
gpt-realtime-2.1-mini