Elon Musk Said Grok 4.7 Would Exceed Every Model. Did It Deliver?
SpaceXAI shipped Grok 4.7, and independent scores place it fourth among frontier AI labs.
Artificial Analysis, an independent benchmarking firm, scored the model at 46 on its Intelligence Index, 2 points above Grok 4.6. Claude Fable 5.1, GPT-6 Astra, and Claude Opus 5 all rank higher.
Grok 4.7 Pushes SpaceXAI Into the Top Four AI Labs
Elon Musk said before launch that the model would run on a 2.1-trillion-parameter base and exceed all current models. He added that the model training included SpaceX company data.
Follow us on X to get the latest news as it happens
But how does Grok 4.7 actually perform? On AA-Briefcase, an Artificial Analysis benchmark for realistic professional work tasks, Grok 4.7 scored 1657 Elo. That marks an 111-point jump over Grok 4.6 and brings the model’s level close to Claude Opus 5.
On GDPval-AA, the model reached 1695 Elo, which puts it 90 points ahead of its predecessor. Coding results moved in the same direction.
Paired with Grok Build, its own coding agent, Grok 4.7 scored 56 on the Coding Agent Index. That is nine points above Grok 4.6.
The model passed GPT-5.6 Sol and now trails only Anthropic’s Claude Fable 5.1, GPT-6 Astra, and Opus 5.
“Grok 4.7 scores 46 on the Artificial Analysis Intelligence Index to bring SpaceXAI into the top 4 AI labs. Coding Agent Index performance has also improved, overtaking GPT-5.6 Sol,” the post read.
Elsewhere, the model barely moved. It gained 4.5 percentage points on Terminal-Bench 4.0 and 3 points on GDP.pdf. Scores slipped on the AA-LCR and AutomationBench-AA tests. Grok 4.5 topped AutomationBench in July, which backed Musk’s claim that it matched Claude Opus.
The Gains Come With a Token Bill
The rate card did not change. Grok 4.7 still costs $2 per million input tokens and $6 per million output tokens. Cache hits stay discounted to $0.50, and the 500,000-token context window carries over from Grok 4.6.
The model does considerably more work for each answer, however. Artificial Analysis measured roughly 81,000 output tokens per Index task, against 36,000 for Grok 4.6 and 27,000 for GPT-6 Astra. Therefore, steady per-token pricing still produces a higher bill per task.
That appetite for tokens lands on a division that has already reported a $1.26 billion loss. Meanwhile, Musk said on September 14 that Grok 4.8 would finish training within a week.
He also expects Grok 5 to be the model that reaches artificial general intelligence. The next release will test whether SpaceXAI can climb the rankings without burning additional compute to do so.
Subscribe to our YouTube channel to watch leaders and journalists provide expert insights
The post Elon Musk Said Grok 4.7 Would Exceed Every Model. Did It Deliver? appeared first on BeInCrypto.
Read moreLatest News
