OpenAI: GPT-3.5 Turbo 16k

Text input Text output
Author's Description

This model offers four times the context length of gpt-3.5-turbo, allowing it to support approximately 20 pages of text in a single request at a higher cost. Training data: up to Sep 2021.

Key Specifications
Cost
$$$$$
Context
16K
Parameters
175B
Released
Aug 27, 2023
Speed
Ability
Reliability
Supported Parameters

This model supports the following parameters:

Logit Bias Tool Choice Seed Top P Top Logprobs Temperature Response Format Logprobs Max Tokens Presence Penalty Tools Frequency Penalty Stop
Features

This model supports the following features:

Response Format Tools
Performance Summary

OpenAI's GPT-3.5 Turbo 16k, released on August 27, 2023, is a robust model offering an extended context length of 16385 tokens, four times that of its predecessor, enabling it to process approximately 20 pages of text. This model consistently performs among the fastest, ranking in the 97th percentile across all benchmarks. While its pricing is moderate, placing it in the 23rd percentile, it demonstrates exceptional reliability, achieving a perfect 100th percentile, indicating minimal technical failures and consistent response delivery. In terms of benchmark performance, the model shows varied strengths. It achieved 99.0% accuracy in Ethics, demonstrating strong moral reasoning, and 95.5% in General Knowledge, indicating a broad understanding of diverse topics. However, its Reasoning accuracy was 50.0%, suggesting a notable weakness in complex multi-step problem-solving. Despite its high speed ranking overall, individual benchmark durations were relatively long, particularly for General Knowledge (151465ms). Its key strengths lie in its extended context window, high reliability, and strong performance in ethics and general knowledge, making it suitable for applications requiring extensive text processing and dependable output.

Model Pricing

Current Pricing

Feature Price (per 1M tokens)
Prompt $3
Completion $4

Price History

Available Endpoints
Provider Endpoint Name Context Length Pricing (Input) Pricing (Output)
OpenAI
OpenAI | openai/gpt-3.5-turbo-16k 16K $3 / 1M tokens $4 / 1M tokens
Azure
Azure | openai/gpt-3.5-turbo-16k 16K $3 / 1M tokens $4 / 1M tokens
Benchmark Results
Benchmark Category Reasoning Free Executions Accuracy Cost Duration
Other Models by openai