gpt-3.5-turbo
Legacy GPT model for cheaper chat
Modalities
In / Out Price
40% off$0.3/ 1M
$0.5 Official price
$0.9/ 1M
$1.5 Official price
Context
16.385K
Released
Sep 21, 2026
API call examples
POST /v1/chat/completionscurl --request POST 'https://api.tokatlas.com/v1/chat/completions' \
--header "Authorization: Bearer $UNIROUTER_API_KEY" \
--header 'Content-Type: application/json' \
--data '{
"model": "gpt-3.5-turbo",
"messages": [
{
"role": "user",
"content": "Explain quantum entanglement."
}
],
"temperature": 0.7
}'Providers & routes
Current provider · OpenAI
Platform configured price · Group: default
USD / 1M tokens. Each row shows input and output from the same tier. Displayed rates use the lowest available group multiplier; your API key group may differ.
GPT-3.5 Turbo as a legacy text baseline
GPT-3.5 Turbo is a legacy OpenAI text model. Use this entry for historical integration references and older prompts. OpenAI shut down its own API access to this model in October 2026. Availability through an existing platform route depends on that platform.
Short summaries
Preserve dates, amounts, and next steps when condensing text.
Text classification
Define labels with clear criteria and a few borderline cases.
Format checks
Use existing parser rules to check missing values, truncation, and invalid output.
Migration baseline
Keep prompts and expected outputs to evaluate an available replacement.
About
Legacy chat interface
GPT-3.5 Turbo is a legacy Chat Completions text model with a historical window of 16,385 tokens and output cap of 4,096. Native input and output are text. Established uses included support summaries, classification, and format conversion.
Retirement and historical results
OpenAI shut down the official API model on October 23, 2026. If a hosted route still uses this string, verify what it resolves to. Historical runs are useful for comparing existing prompts and parsing behavior.
Migrating existing applications
Keep role messages, output contracts, truncation examples, and a small labeled set. Run a replacement through the same parser and check amounts, dates, missing fields, and categories before verifying model and endpoint availability.
Keep a human-verified sample of short and long inputs when evaluating replacements. If a candidate tends to answer at greater length, test application limits and timeouts beyond the simplest cases.
gpt-3.5-turbo FAQ
How do I call this model through the API?
Click Create API Key in the API calls section, or find the button in the Try this model panel at the top of the page. If you’re signed out, sign in to continue to the API keys page. After creating a key, copy the model ID from the header and follow the endpoint and request example.
Where can I see the supported inputs and outputs?
The modality icons at the top show the supported input and output types. If this page has a playground, it also shows the generation options currently available for the model.
How is this model billed?
Check Pricing for billing units and applicable conditions. If the playground shows an estimate, it updates based on the parameters you select.