OpenAI

OpenAI upgraded the flagship reasoning model backing Plus and Pro from GPT-5.6 Sol/Sol Pro to GPT-6 Astra: Plus now lists 'Advanced reasoning models with GPT-6 Astra and GPT-5.6' and Pro's headline reasoning benefit is now 'Pro reasoning powered by GPT-6 Astra'.

Business plan price decreased from $25/user/month (billed annually) to $20/user/month. The 'billed annually' billing qualifier was also removed from the displayed price.

Context window section restructured: simple '16K/32K/128K' table replaced with detailed GPT Instant context window per chat (27K/54K/128K) and GPT Reasoning context window per chat (256K/400K) rows, plus page-equivalent callouts. Plus plan: 'Codex agent and Sora video generation' replaced by 'Expanded Codex usage'. GPT-5.3 Instant upgraded to full access on Go. Sora 1 and Sora 2 rows removed from comparison table. GPT-5 Thinking Mini added as new model row.

Pro plan reasoning model upgraded from GPT-5.2 Pro to GPT-5.4 Pro Pro plan file upload model upgraded from GPT-5.2 to GPT-5.4

Model upgrade: GPT-5.2 Pro replaced with GPT-5.4 Pro across Pro, Business, and Enterprise plans Model upgrade: GPT-5.2 Thinking replaced with GPT-5.4 Thinking across all plans

All public prices removed from ChatGPT pricing page: Free ($0/mo), Go ($8/mo), Plus ($20/mo), Pro ($200/mo), Business ($25/user/mo annually) — prices no longer displayed.

Free plan tagline changed from 'See what AI can do' to 'Intelligence for everyday tasks' Plus plan tagline changed from 'Unlock the full experience' to 'Do more with advanced intelligence'

Pro plan feature description updated from 'Unlimited messages and uploads' to 'Unlimited GPT-5.2 and file uploads', clarifying the model and upload scope. Comparison table feature row renamed from 'Codex agent' to 'Codex'. Plan availability unchanged: Plus=Yes, Pro=Expanded, Business=Yes, Enterprise=Yes.

Atlassian explicitly added to Business plan app integrations list (previously only mentioned Google Drive, SharePoint, GitHub, Notion) New 'Go' plan introduced between Free and Plus tiers with expanded access to GPT-5.2, more messages, uploads, and image creation

GPT-4.1 nano introduced as new ultra-low-cost model priced at $0.100 input / $0.400 output per 1M tokens, optimized for low-latency tasks Primary GPT model input price decreased from $2.50 to $2.00 per 1M tokens (20% decrease) during GPT-4o→GPT-4.1 transition

Assistants API tool pricing removed (Code Interpreter $0.03/session, File Search $0.10/GB/day) Fine-tuning for gpt-3.5-turbo, davinci-002, and babbage-002 models removed

Fine-tuning support added for gpt-4o-mini-2024-07-18 with promotional free tier (2M training tokens daily through October 31, 2024) o1-preview reasoning model introduced with pricing $15.00/1M input tokens and $60.00/1M output tokens (output includes internal reasoning tokens)

GPT-4o model introduced as flagship model at $5.00/1M input tokens and $15.00/1M output tokens Batch API pricing introduced with 50% discount across multiple models

New GPT-4 Turbo model variant gpt-4-0125-preview added with same pricing as existing preview models ($10/1M input, $30/1M output) Pricing display unit changed from 'per 1,000 tokens' to 'per 1M or 1K tokens' with toggle option