SpaceXAI's Grok 4.6 Scores 61 on the Artificial Analysis Intelligence Index
artificialanalysis.ai
August 12, 2026
4 min read
🔥🔥🔥🔥🔥
58/100
Summary
SpaceXAI's Grok 4.6 achieves a score of 61 on the Artificial Analysis Intelligence Index, matching the performance of GPT-5.6 Sol and demonstrating improved agentic capabilities at a lower cost. This version marks a 5-point increase over Grok 4.5 and a 23-point increase compared to Grok 4.3, positioning SpaceXAI alongside OpenAI in the intelligence frontier, just behind Anthropic.
Key Takeaways
- Grok 4.6 scores 61 on the Artificial Analysis Intelligence Index, placing it alongside GPT-5.6 Sol and just behind Claude Opus 5 and Claude Fable 5.
- Grok 4.6 achieves a GDPval-AA v2 Elo of 1753, ranking it behind only Claude Opus 5 and demonstrating strong performance in multi-turn customer service and terminal-based software tasks.
- The pricing for Grok 4.6 remains unchanged at $2/$6 per 1M tokens, offering competitive performance at over 60% lower cost than Claude Opus 5 and GPT-5.6 Sol.
- Grok 4.6 completes long-horizon tasks in approximately 53 turns and 0.5B input tokens, significantly more efficient than Claude Opus 5, which requires around 103 turns and 2.0B input tokens.
Community Sentiment
Positives
- Grok 4.6 is reportedly 3x faster than Claude, which is a game changer for engineers who prioritize speed in their workflow.
- Users appreciate Grok's solid performance and unique speaking style, making it a preferred choice for some despite competition.
- Cursor's subscription model with Grok offers incredible value, allowing extensive use of frontier-level models without running out of tokens.
Concerns
- There's a clear sentiment against Grok, with many stating they wouldn't touch it, regardless of its performance or price.
- Some users express skepticism about Grok's capabilities, suggesting it doesn't advance frontier math and is just decent for basic tasks.
- Concerns about pricing increases for Grok 4.6, especially around cache read costs, raise doubts about its cost-effectiveness for heavy users.
