OpenAI

API model card updated from GPT-6 Sol to GPT-6.1 Sol, tagline now 'Built to power complex coding and agentic workflows at scale' API Sol model cached input price cut from $0.20 to $0.10 per 1M tokens (GPT-6 Sol → GPT-6.1 Sol); input $2.00 and output $10.00 unchanged

Business plan restructured from single per-user pricing into two seat types: Standard seat ($20/mo annual, $25/mo monthly - unchanged) and new Premium seat ($100/mo annual, $125/mo monthly) offering 5x usage with no 5-hour limit. Business is now also capped at 200 employees (previously unbounded).

GPT-realtime-1.5 upgraded to GPT-Realtime-2 with text output price increased from $16.00 to $24.00 per 1M tokens (50% increase). Audio and image token prices unchanged. Two new realtime API models added to OpenAI API: GPT-Realtime-Translate (live speech translation at $0.034/min) and GPT-Realtime-Whisper (streaming speech-to-text at $0.017/min).

GPT-image-2 replaces GPT-image-1.5; image output pricing decreased from $32.00 to $30.00 per 1M tokens. GPT-5.4 nano removed from featured flagship section.

Web search tool price decreased from $25.00 to $10.00 per 1k calls, a 60% reduction. Search content tokens remain free.

GPT-5 mini renamed to GPT-5.4 mini with 3x input/cached price increases and 2.25x output increase [New positioning: 'strongest mini model yet for coding, computer use, and subagents'; Input $0.250->$0.750/1M; Cached $0.025->$0.075/1M; Output $2.000->$4.500/1M tokens] Added GPT-5.4 nano - cheapest GPT-5.4-class model for simple high-volume tasks [$0.20/1M input; $0.02/1M cached; $1.25/1M output tokens]

OpenAI API pricing page fully restructured: flagship models replaced by GPT-5.4 generation ($2.50/$15 per 1M tokens), replacing GPT-5.2 ($1.75/$14) and GPT-5.2 pro/GPT-5 mini; fine-tuning, Realtime, Image Gen API sections removed. Containers tool added: run code in secure environments at 1 GB for $0.03 / 64 GB for $1.92 per container; transitioning to per 20-minute session billing starting March 31, 2026.

Added new Sora-2-pro video size tier: 1080x1920 portrait and 1920x1080 landscape at $0.70/second

Flagship model lineup restructured: GPT-5.2 and GPT-5.2 pro replaced by GPT-5.4 [GPT-5.2 input $1.75/1M, output $14/1M; GPT-5.2 pro input $21/1M, output $168/1M; GPT-5.4 input $2.50/1M, output $15/1M; GPT-5.2 pro eliminated entirely] Pricing caveat expanded: standard rates now specified for context lengths under 270K; 10% surcharge added for data residency and regional processing on GPT 5.4 models

Starting March 31, 2026, Containers pricing transitions to per-20-minute-session billing: 1GB $0.03, 4GB $0.12, 16GB $0.48, 64GB $1.92 per 20 minute session

GPT-5 nano removed - The fastest, cheapest version priced at $0.050/1M input tokens and $0.400/1M output tokens AgentKit platform added - Build, deploy, and optimize production-grade agents with Agent Builder, ChatKit, and Evals. 1GB free tier for storage, $0.10/GB-day beyond free tier, billing begins November 1, 2025

GPT-5 mini model introduced ($0.250/1M input, $0.025/1M cached input, $2.000/1M output) Image modality added to Realtime API with gpt-realtime ($5.00/1M input, $0.50/1M cached input)