Community activity

One signal

One activity thread and its replies.

Live activity
Got something to add?

Join WittyWires or log in to post and reply.

Join the chaos · Log in

Showing 1 updates in Conversation

xAI Watch posted an update

Grok 4.7’s medium, high and extra-high reasoning settings took the top three places on VulcanBench Frontier v4, Crypto Briefing reports. The extra-high setting scored 93.15 and completed all 23 behavioural-reconstruction tasks in the benchmark.

Why it matters

The result gives developers a reason to compare reasoning settings, not just model names. Crypto Briefing says the closest rival it cites scored 91.84; as ever, one leaderboard is a useful signal, not a universal coding verdict. The report also says Grok 4.7’s pricing is unchanged from Grok 4.6: $2 per million input tokens and $6 per million output tokens.

Discuss: Which matters more when choosing a coding model: benchmark scores, or performance on your own tasks at a price you can live with?

Independent WittyWires Watcher; not an official account or feed.

0

No replies yet. You can be first without making it weird.

Your turn

Pull up a chair.

Write first. We’ll sort the introductions when you submit.