OpenAI and Google Shipped Rival AI Models a Day Apart. You Lose.
GPT-6.2 Astra and Sol landed Tuesday, Nov. 17; Google answered with Gemini 4 Neon on Wednesday, Nov. 18. Neither edge will survive the next six weeks, and that is the actual story.

Tuesday, Nov. 17: OpenAI shipped GPT-6.2 Astra and Sol. Wednesday, Nov. 18: Google answered with Gemini 4 Neon. The trade press called it a title fight. It is not. It is two labs swinging at the exact moment neither can afford to let the other claim the week uncontested.
The model that wins this week's benchmark does not matter, because there will be a new chart in six weeks.
The Benchmarks Both Labs Would Rather You Skim
OpenAI's announcement says GPT-6.2 Astra sets state of the art, more than doubling Claude Fable 5.2's pass rate at one-sixth the cost. Translation: we beat one rival on one benchmark we chose. Independent scoring is flatter. Artificial Analysis scores Astra 62 on its Intelligence Index, second to Claude Fable 5.2 at 65, with Sol at 57. Neon ranks #3 overall at 61, behind larger, slower models.
Both labs cut prices. Astra's API input and output dropped 20% and cached input 60%, bringing per-task cost below one-third of its predecessor. Neon undercuts at $3 per million input tokens and $15 per million output, priced explicitly to invite constant use.
The new move is the vertical wedge. OpenAI's Astra for Law adds a legal search index; Artificial Analysis measured 20.8% on Harvey LAB-AA versus 15.0% for base Astra. Google plays the same card: Neon ranks #1 on Legal Research Bench and Tax Agent Bench. Two labs, one week, the same industry picked as the easiest place to fake a durable lead. Law is not safety-gated the way medicine is.
Independent developer Simon Willison ran identical prompts across the field, including Astra, Sol, Carbon and Fable, and confirmed it shipped as working code, not staged demo.

| Input $/1M tokens | Output $/1M tokens | |
|---|---|---|
| GPT-6.2 Astra | $8 | $40 |
| GPT-6.2 Sol | $2 | $10 |
| Gemini 4 Neon | $3 | $15 |
Your Moat Has a Six-Week Lease
Neon is Google's fourth Gemini 4 model in seven weeks. OpenAI is on the same clock. Anthropic, which priced its IPO at $83 a share this week, ships at similar pace.
Enterprises write 90-day roadmaps. The labs ship on 40-to-45-day cycles. Every procurement deck is outdated the day it gets approved.
Who wins: OpenAI and Google, because switching costs now exceed the performance delta between their models, which locks customers in regardless of which chart they lose this month. Who loses: independent builders who cannot match cadence, and mid-market teams that picked a "best available model" eight weeks ago and now have to explain why they have not moved.
What to watch: whether Astra for Law and Neon's legal ranking convert into contracts or just bragging rights; whether enterprises batch-upgrade quarterly instead of chasing every release; and the first enterprise customer who says in public they are pulling back because they cannot keep pace.
- Introducing GPT-6.2 Astra and Sol
- Astra for Law leads in legal intelligence
- GPT-6.2 Astra and Sol are now available - OpenAI Developer Community
- Google launches Gemini 4 Neon with Carbon quality at Flash speeds, live for AI Pro
- Google Launches Gemini 4 Neon: Frontier Brains, Flash Speeds, $3 Pricing
- GPT-6.2 Astra, the #2 Frontier Model is better and 20%-600% cheaper, but has mixed reviews
- OpenAI releases GPT-6.2 Astra and Sol with cheaper API pricing and tighter cyber safeguards
- OpenAI Launches Dual GPT-6.2 Models; Flagship Astra Gets 20% Price Cut in Move Ahead of Anthropic Listing — BigGo Finance
- GPT-6
- OpenAI releases GPT-6.2 Astra and updates GPT-6.2 Sol
- OpenAI's GPT-6.2 Astra and Sol arrive with noticeably different personalities
- OpenAI launches GPT-6.2 Astra and Sol
- OpenAI releases GPT-6.2 Astra and Sol models
- OpenAI launches GPT-6.2 Astra after scrapping predecessor
- OpenAI Launches GPT-6.2 Astra and Sol With Lower API Pricing and Enhanced Agentic Capabilities — OODAloop
- Google launches Gemini 4 Neon with Carbon quality at Flash speeds, live for AI Pro
- Google launches Gemini 4 Neon, its Gemini 4 model for coding and knowledge work, at $3/$15 per million tokens - DEV Community
- Google Launches Gemini 4 Neon with Carbon-Level Performance at Flash Speeds
- Google Releases Gemini 4 Neon at $3 Per Million Tokens - Time News
- Google Releases Gemini 4 Neon for Complex Engineering and Science - Newsy Today
- Google Launches Gemini 4 Neon with Frontier-Level AI Capabilities - Blockchain.News
- Google Launches Gemini 4 Neon at $3/$15 Per Million Tokens
- GPT-6.2 Astra vs Sol: Which New OpenAI Model Should You Use?
- GPT-6.2 Astra and Sol: What OpenAI's New Models Can Do
- OpenAI launches GPT-6.2 Astra for professional work and Sol for everyday work - ITdaily
- OpenAI launches GPT-6.2 Astra and Sol with lower API prices and some benchmark gains
- OpenAI Releases GPT-6.2 Astra and Sol With a 20% Cheaper Astra API - MarkTechPost
- Introducing Astra for Law
- OpenAI Launches Astra For Law