GPT-Realtime-2.1

GPT-Realtime-2.1: a gpt model from OpenAI, ~128K context, knowledge cutoff 2024-09-30

Published

scheduleReleased：July 6, 2026

GPT-Realtime-2.1 is a gpt model from OpenAI (~128K context, input around $4/1M tokens). Suitable for assistants, content generation, knowledge Q&A and business automation

starsCapabilities

visibilityVision understandingcodeFunction calling

paymentsContext and pricing

Context limit128,000

Max output32,000

Knowledge cutoff2024-09-30

Input price$4/ 1M tokens

Output price$24/ 1M tokens

Cached input price$0.4/ 1M tokens

descriptionOverview

Overview

GPT-Realtime-2.1 is provided by OpenAI, model ID gpt-realtime-2.1

Key specs

Context: 128K tokens
Max output: 32K tokens
Knowledge cutoff: 2024-09-30
Input price: $4/1M
Output price: $24/1M

Best for

Consider GPT-Realtime-2.1 when comparing context length, pricing, multimodal support and relay availability

lightbulbUse cases

Image and multimodal Q&A
Visual content analysis
Document and screenshot understanding

thumb_upStrengths

Large context window (~128K)
Supports image and multimodal input
High single-response output limit (~32K)

infoLimitations

Knowledge cutoff 2024-09-30; newer facts need external retrieval
Pricing and availability vary by upstream and relay; verify with official docs and tests

linkReferences

open_in_newhttps://platform.openai.com/docs/models