Grok-1.5 can understand longer documents, handle complex prompts, and shows advancements in reasoning.
Grok-1.5 scored higher than GPT-4 on the HumanEval benchmark, with a score of 74.1% compared to 67% for GPT-4.
Grok-1.5 performed close to Google Gemini on the MMLU test with a score of 81.3%.
Grok 2 is 'in training now' according to Elon Musk, who expects the next generation to outperform existing AI 'on all metrics.'
[
add
]
[
|
|
...
]