US EditionPublished • loading... • Updated
OpenAI Rolls Out GPT-5.2 to Paid ChatGPT Users and API, With Enterprise Access Through Microsoft Foundry
OpenAI's GPT-5.2 series improves accuracy, coding, and long-context understanding, outperforming professionals in 70% of tasks, with a phased rollout starting for paid users.
- On Thursday, OpenAI launched GPT-5.2, rolling it out to ChatGPT paid users and developers via the API in three flavors: Instant, Thinking, and Pro.
- CEO Sam Altman issued a 'code red' memo urging a ChatGPT shift amid traffic declines and Google competition, but OpenAI executives said GPT-5.2 was under months of prior development.
- On benchmarks, GPT-5.2 registered new marks on GDPval across 44 occupations and SWE-Bench Pro at 55.6%, with Pro scoring 93.2%, Thinking solving 40.3% on Tier 1-3 problems, and featuring a 400,000-token context window with a 128,000 max output token limit.
- OpenAI is targeting developers and tooling ecosystem, with steep API pricing and $1.4 trillion in infrastructure commitments raising cost concerns.
- Despite shipping without image upgrades, OpenAI said `On image Gen, nothing to announce today, but more to come` and plans a January model, Adult Mode next year, and Project Garlic in 2026.
Insights by Ground AI
Podcasts & Opinions
192 Articles
192 Articles
ChatGPT Has Seen a Blitz of New Releases Lately. Here's Another.
OpenAI CEO Sam Altman’s “Code Red” declaration earlier this month seemed fairly mild in the face of the dramatic language used by Altman to reportedly fret about Google Gemini’s ascendance in an internal, company-wide memo. ChatGPT 5.2, which launched to users on December 11, has been hyped as, ostensibly, a response to Gemini and fallout from the Code Red memo. AI updates are planned fairly far in advance, and I’d wager since before that memo w…
·New York, United States
Read Full Article
Three of its versions will be available to paid users.
Coverage Details
Total News Sources192
Leaning Left23Leaning Right21Center32Last UpdatedBias Distribution42% Center
Bias Distribution
- 42% of the sources are Center
42% Center
L 30%
C 42%
R 28%
Factuality
To view factuality data please Upgrade to Premium
