June 24, 2026
Smart, cheap, and already starting fights
Grok Build 0.1: Intelligence, Performance and Price Analysis
Grok shocks the comment section as fans, skeptics, and Elon-watchers clash
TLDR: Grok Build 0.1 posted strong scores for quality, speed, and price, making it look like a real contender in the crowded AI race. But the bigger story is the split reaction: some users think it could become a powerhouse, while others refuse to use it because of Elon Musk or complaints about the product experience.
Grok Build 0.1 just dropped some eyebrow-raising numbers: it scores well above average on intelligence tests, runs faster than most rivals, and comes in at a price that looks surprisingly cheap next to other big-name AI models. On paper, it’s a strong showing — especially the low output price — but the real fireworks weren’t in the charts. They were in the comments, where the mood swung wildly between "uh-oh, this is serious" and "absolutely not, I’m not touching it."
One camp sounded genuinely rattled. A few commenters admitted they were surprised by how good Grok felt in real projects, with one practically sounding the alarm that Elon Musk could bulldoze his way into dominating AI coding tools if this keeps improving. Another framed the bigger story as a talent war, saying the rise of top models is proof that some labs are stacked with brilliant people and that kind of success isn’t easy to copy. In other words: this isn’t just about a chatbot — it’s about who gets to rule the next gold rush.
But the backlash was just as loud. Some people said they won’t use Grok on principle because they don’t want to back Musk or his company. Others mocked the product experience, complaining that the wrapping around the model matters as much as the model itself, and that Grok still feels clunky compared with smoother rivals. The running joke? Grok may be smart, fast, and cheap — but if the experience is messy, the internet will roast it anyway.
Key Points
- •Grok Build 0.1 0616 is described as a proprietary reasoning model with text and image input, text output, and a 256k-token context window.
- •The model scores 40 on the Artificial Analysis Intelligence Index, above the comparable-model average of 29.
- •The evaluation reports 130 million generated tokens for the Intelligence Index, compared with an average of 93 million, indicating high verbosity.
- •Pricing is listed at $1.00 per 1M input tokens and $2.00 per 1M output tokens, below the cited class averages of $1.50 and $8.00.
- •The model runs at 93 tokens per second, above the reported average speed of 86 tokens per second.