Perplexity's Sonar—built on Llama 3.3 70b—outperforms GPT-4o-mini and Claude 3.5 Haiku while matching or surpassing top models like GPT-4o and Claude 3.5 Sonnet in user satisfaction. At 1200 tokens/second, Sonar is optimized for answer quality and speed.

Sonar significantly outperforms GPT-4o-mini and Claude 3.5 Haiku in user satisfaction. It also surpasses Claude 3.5 Sonnet and nearly matches GPT-4o, doing so at a fraction of the cost and over 10x faster.

Powered by Cerebras inference infrastructure, Sonar delivers answers at blazing fast speeds, achieving a decoding throughput that is nearly 10x times faster than comparable models like Gemini 2.0 Flash.

We optimized Sonar across two critical dimensions that strongly correlate with user satisfaction — answer factuality and readability. Our results show Sonar outperforms Llama 3.3 70B Instruct and other frontier models in key areas.

@perplexity_ai Sounds like Sonar is finely tuned for literary races next to the creator Titans! Interesting.

@perplexity_ai Exceed sonic speed while delivering magic!

@perplexity_ai Sonar sounds like a speedy trivia wizard. Impressive!

@perplexity_ai @DeepLearn007 @DeepLearn007, exciting advancements in AI technology are on the horizon. 🚀