ModelPickr

GPT-Realtime-2.1 from OpenAI: $4.00 input and $24.00 output per 1M tokens, 128,000 context. Realtime speech-to-speech model with configurable reasoning, tool use, and robust voice-agent behavior Data last refreshed Oct 14, 2026, 05:43 PM UTC.

Models / OpenAI / GPT-Realtime-2.1

OpenAI · gpt

GPT-Realtime-2.1

Realtime speech-to-speech model with configurable reasoning, tool use, and robust voice-agent behavior

$4.00 in · $24.00 out

per 1M tokens · first-party price. Reference price from OpenAI, the lab's own listing.

Context
128,000
Max output
32,000
Cache read
$0.40
Cache write
—
Reasoning
Yes
Tool calls
Yes
Structured output
No
Vision
Yes
Open weights
No
License
—
Released
Jul 6, 2026
Updated in source
Jul 6, 2026
Knowledge
2024-09-30
Providers
2

What this model costs per month

Prices are USD per 1M tokens. A month is 30 days. Cache hits use the cache-read rate when the model has one; otherwise the cache field is ignored.

monthly = requests/day × 30 × (input tokens × (1 − cache hit) × input price + input tokens × cache hit × cache-read price + output tokens × output price) ÷ 1,000,000

Providers

Base rates from Models.dev. One row per provider.
ProviderInputOutputCache readCache writeStatus
OpenAIreference$4.00$24.00$0.40——
Vercel AI Gateway$4.00$24.00$0.40——