Skip to content
TheChatArchiveThe Chat Archive
Perplexitysonar-reasoning-pro-high· PerplexityHugging Face LMArena Search Arena 24K

eres sonar-reasoning-pro-high ?

From a Chatbot Arena battleThe same person sent this prompt to two anonymous models at once. Compare it with Perplexitysonar-pro’s answer.
  1. User

    eres sonar-reasoning-pro-high ?

  2. Perplexity Assistant

    Based on the search results provided, there is no specific product or model called "sonar-reasoning-pro-high" mentioned in the documentation from Perplexity.

    The search results do mention several related Perplexity products:

    1. Sonar Reasoning Pro - A premier reasoning model powered by DeepSeek R1 with Chain of Thought capabilities, created on March 7, 2025, featuring:

      • 128,000 token context window
      • 2/Minputtokensand2/M input tokens and 8/M output tokens pricing
      • Advanced CoT reasoning with real-time internet search and citations
      • US-based hosting with uncensored capabilities
      • Data privacy (no training on customer data)
      • In-depth answers with 2x more citations than standard Sonar Reasoning[3]
    2. Sonar Pro - A premier search offering with higher capabilities than standard Sonar:

      • 200k token context window
      • 3.00per1Minputtokensand3.00 per 1M input tokens and 15.00 per 1M output tokens
      • MMLU score of 0.755 and Intelligence Index of 43
      • Output speed of 75.9 tokens per second
      • Time to first token (TTFT) of 2.59s[1]
    3. Sonar Reasoning - A lighter-weight offering:

      • 127k/130k token context window
      • 1.00per1Minputtokensand1.00 per 1M input tokens and 5.00 per 1M output tokens
      • Output speed of 83.3/83.6 tokens per second
      • Lower latency with 1.81s TTFT[4]

    While there are "high" parameter versions mentioned for other models in the comparison sections (like "o3-mini (high)"), no "high" variant of Sonar Reasoning Pro is explicitly mentioned in these search results.

Responses were produced in different, uncontrolled contexts (system prompts, settings, prior turns). This is not a benchmark.

Identical first prompt