Perplexity: Llama 3.1 Sonar 70B Online

Text input Text output Unavailable
Author's Description

Llama 3.1 Sonar is Perplexity's latest model family. It surpasses their earlier Sonar models in cost-efficiency, speed, and performance. This is the online version of the [offline chat model](/models/perplexity/llama-3.1-sonar-large-128k-chat). It is focused on delivering helpful, up-to-date, and factual responses. #online

Key Specifications
Cost
$$$$
Context
127K
Parameters
70B
Released
Jul 31, 2024
Speed
Ability
Reliability
Supported Parameters

This model supports the following parameters:

Top P Max Tokens Presence Penalty Temperature Frequency Penalty
Performance Summary

Perplexity's Llama 3.1 Sonar 70B Online model demonstrates moderate speed performance, ranking in the 30th percentile across various benchmarks, indicating it performs at an average pace compared to other models. In terms of cost, it offers competitive pricing, placing in the 41st percentile, making it a reasonably economical option. Analyzing its performance across specific benchmarks reveals a mixed profile. The model excels in Email Classification, achieving 97.0% accuracy, placing it at the 50th percentile for this task, suggesting a strong capability in categorizing structured text. Its Reasoning abilities are fair, with 48.0% accuracy, though its speed in this area is notably slow (13th percentile). However, the model exhibits significant weaknesses in other areas. Its performance in Coding (Baseline) is very poor, with only 1.0% accuracy, ranking it at the 8th percentile. Similarly, its General Knowledge is limited, achieving only 8.5% accuracy (14th percentile). While its Ethics performance is 81.0% accurate, this places it in the lower 23rd percentile, indicating room for improvement in nuanced ethical scenarios. Overall, the Llama 3.1 Sonar 70B Online appears to be a cost-effective option for specific classification tasks but struggles considerably with complex coding, general knowledge, and advanced reasoning, despite its "helpful, up-to-date, and factual responses" description.

Model Pricing

Current Pricing

Feature Price (per 1M tokens)
Prompt $1
Completion $1
Request $0.005 per request

Price History

Available Endpoints
Provider Endpoint Name Context Length Pricing (Input) Pricing (Output)
Perplexity
Perplexity | perplexity/llama-3.1-sonar-large-128k-online 127K $1 / 1M tokens $1 / 1M tokens
Benchmark Results
Benchmark Category Reasoning Free Executions Accuracy Cost Duration
Other Models by perplexity