AI Hub
← All models

Llama 3 70B Instruct vs Sonar Reasoning Pro

Meta vs Perplexity — benchmarks, pricing, and capabilities side by side.

  • Sonar Reasoning Pro has the higher intelligence index (95.7 vs 40.8)
  • Llama 3 70B Instruct is cheaper ($0.51 vs $2.00 per 1M input)
  • Sonar Reasoning Pro has a larger context window (128K)
Llama 3 70B InstructSonar Reasoning Pro
Intelligence index40.895.7
DeveloperMetaPerplexity
TypeLLMMultimodal
AccessOpen weightsAPI only
Context window8,192 tokens128,000 tokens
Input price$0.51 / 1M$2.00 / 1M
Output price$0.74 / 1M$8.00 / 1M
Speed45 tok/s
ReleasedApril 18, 2024March 7, 2025
Parameters
Input modalitiesTextText, Image
Output modalitiesTextText

Shared benchmarks

Llama 3 70B Instruct
Sonar Reasoning Pro
MATH-500
48.3
95.7