grok 4.7 really likes to think.
This is a AI post classified by Jev as Frontier models (a model release), kept by the AI Radar because it carries real work, not commentary.
grok 4.7 really likes to think. at xhigh, grok 4.6 used ~20k reasoning tokens per artificial analysis intelligence index task. grok 4.7 uses ~59k. that's 3x more reasoning tokens over a generation.
Posted by Ananth (3.4k followers) 57 min ago · 16 likes · 551 views · view the original post on X. Kept by the AI Radar as Frontier models.
More AI work like this
- You can use the new Grok 4.7 model with similar-level performance, 83% cheaper than… — @0x0SojalSec
- BREAKING: SpaceXAI just dropped Grok 4.7 — @ns123abc
- Grok 4.7 is 83% cheaper than GPT-6 Astra & 88% cheaper then Fable 5.1. — @0x0SojalSec
- They are comparing grok 4.7 xhigh with grok 4.6 high — @chetaslua
- my first test with Grok 4.7 fast high in Grok Build — @HarshithLucky3
- Grok 4.7 Is 8x Cheaper Than GPT-6 Astra And Still Competitive 💰 — @Priyannkaaaa
- Grok 4.7 is live in Droid. Observations so far: — @FactoryAI
- We evaluated Grok 4.7 across the Vals benchmark suite. It ranks #24 on the Vals Index at… — @ValsAI
Every post is read and classified by Jev (TypeSafe): what it is, which market it belongs to, and whether the link is a real tool. 39.4k posts from 5k X accounts over the last 14 days, 1.6k tools, 19 markets. Collected every 5 minutes, fully re-ranked every hour — last update 2026-09-21 17:38 UTC. Full method.