OpenAI: GPT-4
The OpenAI GPT-4 is a language model that handles text input and output, with a context length of 8191 tokens. It supports tools but does not have reasoning capabilities. The structured output is not available. Those considering the GPT-4 for their use case should shortlist it if they require a text-based solution with long context lengths. Its price point of $30 per million input tokens and $60 per million output tokens may be reasonable for those who can utilize its capabilities within these limits. However, it ranks lower in benchmarks compared to other models, with a blended score of 6.0 across one independent benchmark.
Benchmark results
Independent, published benchmarks. Blended score 5.5 across 1 benchmark, last refreshed 2026-10-13. How scoring works →
| Benchmark | Measures | Score |
|---|---|---|
| AI Index | broad capability composite | 11.1 |
- Model ID
- openai/gpt-4
- Vendor
- openai
- Released
- May 2023
- Tokenizer
- GPT
- Input Modalities
- text
- Output Modalities
- text
- Max Output
- 4,096 tokens
- Tool Calling
- ✓ supported
- Structured Output
- ✓ supported
- Reasoning Mode
- not supported
- Vision
- text only
- Audio
- no
- Moderated
- no
What it costs in practice
Computed from the current $30.00/M input and $60.00/M output rates. Run your own numbers →
| Job | Tokens | Cost |
|---|---|---|
| Summarize a 50-page report | 30k in / 1.5k out | $0.99 |
| Classify 1,000 customer emails | 500k in / 50k out | $18.00 |
| A month of a busy support chatbot | 5M in / 2M out | $270.00 |
Similar models
OpenAI: GPT-3.5 Turbo Instruct
OpenAI: GPT-3.5 Turbo 16k
OpenAI: GPT-3.5 Turbo (older v0613)
OpenAI: GPT-4 Turbo
OpenAI: GPT-3.5 Turbo
OpenAI: GPT-3.5 Turbo (batch)
Quick answers
- How much does OpenAI: GPT-4 cost?
- $30.00 per million input tokens and $60.00 per million output tokens.
- What is OpenAI: GPT-4's context window?
- 8,191 tokens, roughly 12 pages of text in a single request.
- Does OpenAI: GPT-4 support tool calling?
- It supports tool calling, structured output.
- Can OpenAI: GPT-4 process images?
- No, it is text-only on the input side.