GPT-5.4 Mini
GPTBalanced
ThinkingTool UseVisionStructured Output
About this model
GPT-5.4 mini brings the core capabilities of GPT-5.4 to a faster, more efficient model optimized for high-throughput workloads. It supports text and image inputs with strong performance across reasoning, coding,...
Performance Tier
Balanced
GPT-5.4 Mini is a balanced model from GPT : strong performance at a reasonable price.
Strong cost-performance ratio. Reliable for most professional use cases without premium pricing.
Pricing
This model is included in Elosia plansModerate
Moderate cost. A balanced choice for regular use without constant cap watching.
| Type | per 1M tokens |
|---|---|
| Input (prompt) | $0.750 |
| Output (completion) | $4.50 |
| Cache read | $0.075 |
Capabilities
Context Length400K
Max Output Tokens128K
TokenizerGPT
Inputfile, image, text
Outputtext
Release DateMarch 17, 2026
Benchmarks
General Intelligence
MMLU
Not reported
Mathematics
MATH-500
Not reported
Programming
HumanEval
Not reported
Reasoning
Humanity's Last Exam
28.2%
Agentic
SWE-bench Pro
54.4%
Recommended Use Cases
CodingAnalysisData ExtractionGeneral ChatSummarization
Strengths
- Approaches GPT-5.4 performance at a fraction of the cost ($0.75/M vs $2.50/M input)
- 2x faster than GPT-5 mini with improvements across coding, reasoning, and tool use
- Native computer use and multimodal capabilities inherited from GPT-5.4
- 400K context window with native tool use, web search, and function calling
Limitations
- Smaller context window than GPT-5.4 (400K vs 1.05M tokens)
- Limited reasoning depth compared to the full GPT-5.4 model
Resources
This model may use your data for training