EmbeddingGemma 2: an open, lightweight multimodal embedding model
deepmind.google·6d ago
AI Engineering
Hacker News·Aug 12·5 min read
SpaceXAI's Grok 4.6 Scores 61 on the Artificial Analysis Intelligence Index
TL;DR
SpaceXAI's Grok 4.6 has achieved a score of 61 on the Artificial Analysis Intelligence Index, marking a significant improvement over its predecessor and positioning it alongside leading models like GPT-5.6 Sol.
✦ Why It Matters
Consider integrating Grok 4.6 into your projects for advanced AI capabilities at a lower cost.
Key Takeaways
- Grok 4.6 scores 61 on the Intelligence Index, improving agentic performance.
- Pricing remains competitive at $2/$6 per million tokens.
- Excels in long-horizon tasks with efficient token usage.
- Consider Grok 4.6 for cost-effective AI solutions in your applications.
Full Summary
Grok 4.6, developed by SpaceXAI, has scored 61 on the Artificial Analysis Intelligence Index, reflecting a 5-point increase from its predecessor, Grok 4.5. This model demonstrates strong agentic performance, particularly in real-world tasks, with an Elo score of 1753, placing it just behind Claude Opus 5.
Notably, Grok 4.6 maintains its pricing at $2/$6 per million tokens, significantly lower than competitors like Claude Opus 5 and GPT-5.6 Sol. It excels in long-horizon knowledge work, completing tasks in about 53 turns and 0.5 billion input tokens, which is more efficient than Claude Opus 5.
The model's performance across various benchmarks indicates its versatility in knowledge work, customer service, and terminal-based tasks, making it a strong contender in the AI landscape.