header-langage
简体中文
繁體中文
English
Tiếng Việt
한국어
日本語
ภาษาไทย
Türkçe
Scan to Download the APP
Partnered With:

Musk's Latest AI Release Market Expected to Underperform Gemini 3 Pro

According to PolyBeats monitoring, on the prediction market Polymarket, the current probability of the market "Did the next Grok model released by xAI score at least 1480 on debut?" being "Yes" is 22%.

On July 8th, xAI officially released Grok 4.5, touting it as the company's strongest model to date, focusing on code, Agentic tasks, and knowledge work, and has been made available on Grok Build, Cursor, and the xAI API console. xAI stated that Grok 4.5 was trained on tens of thousands of NVIDIA GB300 GPUs, strengthened on software engineering, mathematical, scientific, and engineering data, emphasizing its inference efficiency, 80 TPS output speed, and lower token cost.

The 1480 score threshold of this market is close to the top model range: xAI previously claimed its Thinking version achieved 1483 Elo in the Text Arena when it released Grok 4.1 last year. The current "Yes" probability of scoring 1480+ on Polymarket is only 22%, with the score of Gemini 3 Pro, currently ranked on the public leaderboard, at around 1490 Elo, belonging to the top range.
---------------------------------
See tomorrow today, follow @PolyBeatsEN to get a glimpse of the future with @PolyBeats_Bot!

举报 Correction/Report
Correction/Report
Submit
Add Library
Visible to myself only
Public
Save
Choose Library
Add Library
Cancel
Finish