Megadose AI progress, ranked and analyzed.

Grok 4.6 scores 61 on the Artificial Analysis Intelligence Index

· HN · Frontpage AI ·
Artificial Analysis puts Grok 4.6 back on the frontier mainly on agentic work and cost efficiency.

The piece says Grok 4.6 gains 5 index points over Grok 4.5 and lands alongside GPT-5.6 Sol, behind only Claude Opus 5 and Claude Fable 5. Its strongest showing is in agentic benchmarks, including GDPval-AA v2, 𝜏³-Banking, and Terminal-Bench v2.1. Artificial Analysis says its $2/$6 per 1M token pricing is unchanged from Grok 4.5 and well below Claude Opus 5 and GPT-5.6 Sol. On AA-Briefcase, it scores below the Claude Opus 5 family but uses fewer turns and fewer input tokens on average. HN · Frontpage AI's note

score 7

Categories: Model Releases

Discussions

  • hn · 259 points · 248 comments
  • hn · 281 points · 266 comments
  • hn · 288 points · 277 comments
  • hn · 295 points · 283 comments
  • hn · 300 points · 296 comments
  • hn · 305 points · 308 comments