openai

OpenAI: GPT-4o (batch)

OpenAI's GPT-4o is an advanced AI model capable of handling text, images, and files within a context length of 128,000 tokens. It supports tools integration but lacks reasoning capabilities or structured output formats. The model demonstrates versatility in processing multiple input modalities. For users looking for a reliable tool with a balance between cost and performance, GPT-4o is worth considering. Its blended benchmark score of 35.0 across two independent evaluations, along with its moderate pricing at $1.25 per million input tokens and $5.0 per million output tokens, makes it suitable for those who need efficient handling of diverse inputs without the requirement for complex reasoning tasks or structured outputs.

Quality Score
92/100
price + capability + benchmarks
Input Price
$1.25
per 1M tokens
Output Price
$5.00
per 1M tokens
Context Window
128,000
tokens

Benchmark results

Independent, published benchmarks. Blended score 35.0 across 2 benchmarks, last refreshed 2026-08-07. How scoring works →

BenchmarkMeasuresScore
Aider Polyglot code editing across languages 45.3
AI Index broad capability composite 18.3
Model ID
openai/gpt-4o:batch
Vendor
openai
Released
May 2024
Tokenizer
GPT
Input Modalities
text, image, file
Output Modalities
text
Max Output
16,384 tokens
Tool Calling
✓ supported
Structured Output
✓ supported
Reasoning Mode
not supported
Vision
✓ accepts images
Audio
no
Moderated
yes

What it costs in practice

Computed from the current $1.25/M input and $5.00/M output rates. Run your own numbers →

JobTokensCost
Summarize a 50-page report 30k in / 1.5k out $0.04
Classify 1,000 customer emails 500k in / 50k out $0.88
A month of a busy support chatbot 5M in / 2M out $16.25

Similar models

Quick answers

How much does OpenAI: GPT-4o (batch) cost?
$1.25 per million input tokens and $5.00 per million output tokens.
What is OpenAI: GPT-4o (batch)'s context window?
128,000 tokens, roughly 192 pages of text in a single request.
Does OpenAI: GPT-4o (batch) support tool calling?
It supports tool calling, structured output.
Can OpenAI: GPT-4o (batch) process images?
Yes, it accepts image input.