GPT-5.4 mini
GPT-5.4 mini costs $0.75 per 1M input tokens and $4.50 per 1M output tokens. Its context window is 400,000 tokens. GPT-5.4 mini has no announced retirement. Checked 2026-10-21.
"Our strongest mini model yet for coding, computer use, and subagents"
Facts
| API model ID | gpt-5.4-mini | Source |
|---|---|---|
| Vendor | OpenAI | Source |
| Input price | $0.75 per 1M tokens | Source |
| Output price | $4.50 per 1M tokens | Source |
| Cached input price | $0.075 per 1M tokens | Source |
| Context window | 400,000 tokens | Source |
| Max output | 128,000 tokens | Source |
| Inputs | Text, Images | Source |
| Released | Not published | Source |
| Knowledge cutoff | Aug 31, 2025 | Source |
| Retirement | Active. No retirement date listed. | Source |
The default snapshot is gpt-5.4-mini-2026-03-17.
What replaces it, and what it replaces
- OpenAI has not named a replacement.
- For older models that already have shutdown dates, see our AI model retirement dates calendar.
Benchmark scores
Independent scores from Epoch AI, which scores every vendor's models the same way. The rank compares GPT-5.4 mini with the other models on our AI models page.
| Benchmark | GPT-5.4 mini |
|---|---|
| Epoch Capabilities Index One score that combines many benchmarks, such as math, science, and coding tests, into a single scale. Epoch sets Claude 3.5 Sonnet at 130 and GPT-5 at 150. | 148.70 Rank 15 of 17 models here |
| GPQA Diamond 198 hard multiple-choice questions in biology, chemistry, and physics, written by PhD-level experts. Guessing scores 25%. We show the highest score for each model across the reasoning settings Epoch ran. | 86.9% Rank 14 of 16 models here |
Scores are Epoch AI's, not ours. Source: Epoch AI, Capabilities and benchmarking, from its public data download, checked 2026-10-21. Used under CC BY 4.0. We drew the charts. A benchmark is one test, so run your own inputs before you choose.
What it costs for your workload
At 10,000 requests a month of 5,000 input and 500 output tokens, GPT-5.4 mini costs about $60 a month. Change the numbers to match yours.
Monthly cost estimate
Example sizes are our round numbers. Set your own below.
GPT-5.4 mini $60 a month
$6.00 per 1,000 requests
List prices checked 2026-10-21, standard tier, no caching, no batch discount. Claude 4.7 and later count about 30% more tokens for the same text, so their rows include that. Long-prompt prices apply where the vendor sets them.
Compare it
- Compare GPT-5.4 mini with any other model
- Not sure which model fits the job? The AI model picker gives you three to test.
Questions clients ask
Short answers, no sales language. If yours isn't here, ask an engineer directly.
Ask us anythingHow much does GPT-5.4 mini cost?
$0.75 per 1M input tokens and $4.50 per 1M output tokens at list price. At 10,000 requests a month of 5,000 input and 500 output tokens, that is about $60 a month. Checked 2026-10-21.
What is the context window of GPT-5.4 mini?
400,000 tokens, with up to 128,000 output tokens per request.
When will GPT-5.4 mini be retired?
No retirement is announced on OpenAI's deprecation page as of 2026-10-21.
Can GPT-5.4 mini read images or PDFs?
OpenAI lists these inputs: text, images. PDF is not listed as an input type on the model page.
Building on GPT-5.4 mini?
We add models to products and test them on your real inputs. We bill hourly or by milestone.