modelgrep

Best Perplexity Models for RAG

Match · Updated July 2026

The best Perplexity model for RAG is Sonar Reasoning Pro — 17.8 intelligence with 128K tokens of context for retrieved passages. Sonar Pro (9.3) is next.

17.8Intelligence
$2.00Input /M
128KContext

The best AI models for retrieval-augmented generation, ranked by intelligence among models with at least 128K tokens of context — enough to hold retrieved passages plus conversation. RAG pipelines run at volume, so weigh the price and speed columns as hard as the score.

  1. 1perplexity logo
    sonar-reasoning-pro
    ReasoningVision17.8 intel · $2.00/M · 128K ctx
    17.8
    Intelligence
  2. 2perplexity logo
    sonar-pro
    Vision9.3 intel · $3.00/M · 200K ctx
    9.3
    Intelligence

Frequently asked

What is the best Perplexity model for RAG?

The best Perplexity model for RAG is Sonar Reasoning Pro — 17.8 intelligence with 128K tokens of context for retrieved passages. Sonar Pro (9.3) is next.

What's a good alternative to Sonar Reasoning Pro?

Sonar Pro (9.3) is the closest alternative on this metric. See the full ranking above for the tradeoffs.

How many Perplexity models are there?

modelgrep tracks 5 Perplexity models with live benchmarks, speed, latency and per-provider pricing, led on intelligence by Sonar Reasoning Pro. 2 of them qualify for this ranking.

More Perplexity rankings

All rankings