@jolyonGrok 4.6 is out, and it's a major step up from Grok 4.5 without a price change. It posts the top scores on GDPVal, AA-Briefcase, and Harvey LAB, and it holds its own across a wide range of benchmarks against other leading AI models. This sets a new bar.
Orijinal gönderiyi gör@jolyon tarafından fotoğraf· Aug 12, 2026· Grok
The image is a data table comparing performance scores of different AI models. The table has rows listing various benchmarks like "AA Intelligence Index" and "GDPVal-AA v2," and columns showing scores for "Grok 4.6 High," "Grok 4.5 High," "GPT-5.6 Sol Max," and "Fable 5 Max." The style is clean and informative, resembling a technical report or infographic. Notably, some scores are highlighted in bold, suggesting superior performance in those categories. The text at the bottom reads: "Third-party model scores best of self-reported or publicly available."
bee collecting nectar
young boy soccer uniform
elderly man in hospital bed
thank you to Grok
@grok family portrait
2Elon Musk and Chess.com exchange
grok chatgpt gemini claude outage
3ChatGPT Claude Grok outage
3Grok Bot architecture working note
Grok Bot vs API vs CLI explained
2Grok removing anime companions on X
GPT Astra vs Fable 5.1 benchmark
2@grok hairstyle swap
Sydney J. Harris quote on AI
2Billie Eilish perfume ad
grok reaction
2Andrew Garfield bear face edit
@grok skin color change@jolyonGrok 4.6 is out, and it's a major step up from Grok 4.5 without a price change. It posts the top scores on GDPVal, AA-Briefcase, and Harvey LAB, and it holds its own across a wide range of benchmarks against other leading AI models. This sets a new bar.
Orijinal gönderiyi gör@jolyon tarafından fotoğraf· Aug 12, 2026· Grok
The image is a data table comparing performance scores of different AI models. The table has rows listing various benchmarks like "AA Intelligence Index" and "GDPVal-AA v2," and columns showing scores for "Grok 4.6 High," "Grok 4.5 High," "GPT-5.6 Sol Max," and "Fable 5 Max." The style is clean and informative, resembling a technical report or infographic. Notably, some scores are highlighted in bold, suggesting superior performance in those categories. The text at the bottom reads: "Third-party model scores best of self-reported or publicly available."
bee collecting nectaryoung boy soccer uniformelderly man in hospital bedthank you to Grok@grok family portrait2Elon Musk and Chess.com exchangegrok chatgpt gemini claude outage3ChatGPT Claude Grok outage3Grok Bot architecture working noteGrok Bot vs API vs CLI explained2Grok removing anime companions on XGPT Astra vs Fable 5.1 benchmark2@grok hairstyle swapSydney J. Harris quote on AI2Billie Eilish perfume adgrok reaction2Andrew Garfield bear face edit@grok skin color change