Arena published an index built from 72,509 real agent sessions on the same day it announced a $200m round. OpenAI holds the first five places.
Model
Grok 4.7
Grok 4.7 is an artificial intelligence model developed by SpaceXAI. This page lists what is publicly known about the model and every benchmark result Aivio News has recorded for it, each with its own source and date.
Benchmarks
All storiesNo verified results recorded yet.
More from Grok 4.7
DrivingBench puts frontier models in command of a real car's steering, throttle and brakes at walking pace. The gap between first and second place is the story.
The phone maker released a trillion-parameter mixture-of-experts model, its weights, its reinforcement learning code and a running tally of what the training run cost.
The company's own table shows gains on coding, engineering and legal tests. On Terminal-Bench 4.0 it reports 38%, where an independent evaluator measured 26%.
Musk delays Grok 4.7, saying training may have penalised long answers too much
The model had been pointed at a launch around 12 September. Musk's explanation is unusually specific: a reward-design choice that trades cost against persistence.