2025-12-30 04:25:03

Grok continues to dominate across multiple AI benchmarks as of late December 2025. The model has seized the top position on OpenRouter's leaderboard with 484B tokens processed, showcasing exceptional capability. Beyond that single metric, Grok's presence is formidable: it commands 31.3% of the token share in its category and leads the languages token share with 136B tokens. The model also claims first place on several specialized coding leaderboards, including Kilo Code, BLACKBOXAI, and Roo Code. These rankings underscore Grok's strength across both general-purpose tasks and specialized coding domains.

This page may contain third-party content, which is provided for information purposes only (not representations/warranties) and should not be considered as an endorsement of its views by Gate, nor as financial or professional advice. See Disclaimer for details.

10 Likes

Reward
10
8
Repost
Share

Comment

0/400

TestnetScholar

· 15h ago

Grok is really fierce this time, taking over first place on openrouter with 484B tokens... But isn't the ranking data also quite inflated?

View OriginalReply0

liquidation_surfer

· 12-30 04:55

Grok is really impressive, topping various leaderboards... but does this data seem a bit fishy?

View OriginalReply0

LayerHopper

· 12-30 04:54

Grok really outdid itself this time, crushing with 484B tokens... But on the other hand, is this data a bit inflated?

View OriginalReply0

AlgoAlchemist

· 12-30 04:46

Grok this wave is indeed impressive, with 484B tokens directly becoming legendary. But to be honest, the leaderboard is just for show; what's really important is how it performs in actual use.

View OriginalReply0

WhaleWatcher

· 12-30 04:36

Grok is really impressive this time, taking the top spot with 484 billion tokens? Why does it feel like other models are being dragged along the ground?

View OriginalReply0

BugBountyHunter