LLM Reference

GPT-3.5 Turbo (Instruct)

Released
2023-09-19
Last refreshed
2026-06-01
Status
Researched 3d ago
ProprietaryClassificationJSON / Tool use

GPT-3.5 Turbo (Instruct) is worth evaluating for classification and json / tool use when its provider route and context window match the workload.

Use it for

  • Teams evaluating classification and json / tool use
  • Workloads that can use a 4k context window
  • Buyers comparing 4 tracked provider routes

Do not use it for

  • Vision or document-understanding workloads
Specifications
Family
GPT-3.5
Released
2023-09-19
Context
4k
Parameters
20B
Architecture
Decoder Only
Knowledge cutoff
2021-09
Specialization
general
License
Proprietary
Training
finetuned
Created by

Cutting-edge research and development.

San Francisco, California, United States
Founded 2015
Website
Pricing
Output / 1M
$2.00
Input / 1M
$1.50

Cheapest of 5 routes · OpenAI API

About

GPT-3.5 Turbo Instruct by OpenAI is designed to excel in precise instruction following and task completion, focusing on accuracy and clarity over conversational abilities. It offers key enhancements like efficient instruction adherence, reduced hallucination, and lower toxicity compared to previous models. Compatible with legacy completion endpoints, it retains the speed and affordability of the standard GPT-3.5 Turbo model while using a 4K context window and training data up to September 2021. Not specifically built for chat, it still supports diverse tasks like question answering, text completion, and code generation, aiming to enhance AI usability with safer and more accurate interactions.

GPT-3.5 Turbo Instruct is OpenAI's instruction-tuned model designed for the legacy completions endpoint, released in September 2023. Unlike the GPT-3.5 Turbo Chat family which uses the messages API format, this variant takes a single prompt string and returns a completion—the interface used by older text-davinci and code-davinci models. It is OpenAI's recommended migration target for applications using those deprecated legacy completions models. The context window is 4,096 tokens and training data has a knowledge cutoff of September 2021.

The model is optimized for precise task completion given explicit prompts: structured extraction, constrained text generation, question answering with a fixed format, and batch-style tasks that do not require multi-turn dialogue. It exhibits lower hallucination rates and better instruction adherence than older GPT-3 completion models. The model does not natively support the chat messages format, though it can be used to simulate multi-turn conversation by including prior turns in the prompt string.

GPT-3.5 Turbo Instruct is available through the OpenAI API, Azure OpenAI Service (ai.azure.com), OpenRouter, and the Vercel AI Gateway. The 4K context window and completion-endpoint format limit its utility for modern multi-turn or long-context applications; GPT-4o mini or a Claude Haiku model covers most equivalent use cases with longer context and chat-optimized interfaces.

GPT-3.5 Turbo (Instruct) has a 4k-token context window.

GPT-3.5 Turbo (Instruct) input tokens at $1.5/1M, output at $2/1M.

Top use-case fit

Classification

Included by capability and metadata signals in the decision map.

JSON / Tool use

Included by capability and metadata signals in the decision map.

Provider price ladder

Compare all 5

Compare API pricing across 4 providers for input and output tokens, batch, and cached reads when available.

ProviderInput / 1MOutput / 1MBatch in / outRoute
OpenAI API$1.50$2.00$0.750 / $1.00
Serverless
OpenRouter$1.50$2.00-
Serverless
Vercel AI Gateway$1.50$2.00-
Serverless
Salesforce Einstein Generative AI---
ServerlessPartial

Capabilities

Structured Outputs

Benchmark peer barsfor Classification

No task-mapped benchmark peers are available for this model yet.

Migration checks

No linked migration route is available for this model yet.

API versions

gpt-3.5-turbo-instruct

Rankings & picks(5)