BenchMark'd
Based on 2 reviews
Perplexity's most advanced search-augmented reasoning model, built on Llama 3.3 70B with real-time web access. Combines deep reasoning with up-to-date information retrieval for complex research queries.
Released
October 15, 2025
Parameters
Unknown
Context
200K
Pricing
Paid
| Benchmark | Category | Score | Performance |
|---|---|---|---|
MMLU | knowledge | 79.5% | 80 |
Chatbot Arena ELO | overall | 1280 | 69 |
Last updated: March 15, 2026
Benchmark scores may vary based on evaluation methodology and conditions.