AI Benchmark Model
Grok 4.7 (high)
Benchmark scores, pricing, speed, and model comparisons for Grok 4.7 (high).
Key scores
Grok 4.7 (high) is an artificial intelligence model developed by SpaceXAI and released on September 21, 2026. The model holds an intelligence index score of 46.3. Performance metrics for the model include an output speed of 38.606 tokens per second and a time to first token of 0.732 seconds. Regarding benchmark results, the model achieved a score of 0.423 on HLE, 0.578 on SciCode, and 0.77 on LCR. Pricing for Grok 4.7 (high) is set at $2.00 per 1 million tokens for input and $6.00 per 1 million tokens for output, resulting in a blended rate of $3.00 per 1 million tokens. Coding and math index data for this model remain unknown.
Benchmark results
3 tracked benchmark rows for Grok 4.7 (high).
LCR Other
77.0% SciCode Other
57.8% HLE Other
42.3%