OpenAI: GPT-3.5 Turbo (batch)
Openai
GPT-3.5 Turbo is OpenAI's fastest model. It can understand and generate natural language or code, and is optimized for chat and traditional completion tasks. Training data up to Sep 2021.
Pricing
Source-reported prices are shown in their native units. No missing fee is estimated.
Input
$0.250
USD / 1M input tokens
Output
$0.750
USD / 1M output tokens
Capabilities
A check means the source explicitly lists support; a dash means the source did not publish a value.
Text input Supported
Text output Supported
Vision Not listed
Tools Supported
Reasoning Not listed
Structured outputs Supported
Audio Not listed
Image generation Not listed
Embeddings Not listed
Alternatives
Suggestions use deterministic rules: overlapping modalities, a context window within 2×, and shared source-reported signals. No AI-generated recommendations.
Similar models
Mistral: Saba
mistralai/mistral-saba
Input
$0.200
Intelligence
6.5
Similar category, context, price or shared benchmark signals.
OpenAI: GPT-3.5 Turbo
openai/gpt-3.5-turbo
Input
$0.500
Intelligence
5.5
Similar category, context, price or shared benchmark signals.
Qwen2.5 Coder 32B Instruct
qwen/qwen-2.5-coder-32b-instruct
Input
$0.660
Intelligence
6.7
Similar category, context, price or shared benchmark signals.
Amazon: Nova 2 Lite
amazon/nova-2-lite-v1
Input
$0.300
Intelligence
13.4
Similar category, context, price or shared benchmark signals.
Cheaper alternatives
Strictly lower or equal input and output pricing, with no lower shared benchmark value and similar context.
Mistral: Mistral Small 3
mistralai/mistral-small-24b-instruct-2501
Input
$0.050
Intelligence
6.7
Lower source-reported input and output cost while matching or exceeding shared benchmarks.
Microsoft: Phi 4
microsoft/phi-4
Input
$0.070
Intelligence
5.9
Lower source-reported input and output cost while matching or exceeding shared benchmarks.
Mistral: Saba
mistralai/mistral-saba
Input
$0.200
Intelligence
6.5
Lower source-reported input and output cost while matching or exceeding shared benchmarks.
Higher-performance alternatives
Higher on at least one shared benchmark and never lower on the other shared benchmarks, with similar context.
Mistral: Mistral Small 3
mistralai/mistral-small-24b-instruct-2501
Input
$0.050
Intelligence
6.7
Higher source-reported performance on shared benchmark dimensions.
Qwen2.5 Coder 32B Instruct
qwen/qwen-2.5-coder-32b-instruct
Input
$0.660
Intelligence
6.7
Higher source-reported performance on shared benchmark dimensions.
Mistral: Saba
mistralai/mistral-saba
Input
$0.200
Intelligence
6.5
Higher source-reported performance on shared benchmark dimensions.
Microsoft: Phi 4
microsoft/phi-4
Input
$0.070
Intelligence
5.9
Higher source-reported performance on shared benchmark dimensions.