O4 Mini
OpenAI o4-mini is a compact reasoning model in the o-series, optimized for fast, cost-efficient performance while retaining strong multimodal and agentic capabilities. It supports tool use and demonstrates competitive reasoning and coding performance across benchmarks like AIME (99.5% with Python) and SWE-bench, outperforming its predecessor o3-mini and even approaching o3 in some domains.
Input
Output
Context
200K
Max Output
100K
Parameters
—
Technical Specifications
Model TypeChat
Context Window200K
Max Output Tokens100K
Parameters—
Training Cutoff—
Licenseproprietary
Capabilities
Input Modalities
imagetextfile
Output Modalities
text
Features
chain_of_thoughtcode_generationextended_thinkingfunction_callingreasoningvision
Benchmarks
17.5%
HLE
94.0%
AIME
78.4%
GPQA
46.5%
SciCode
98.9%
MATH 500
83.2%
MMLU Pro
90.7
Math
25.6
Coding
85.9%
LiveCodeBench
33.1
Intelligence
129
Speed (tok/s)
15.6
TTFA (s)
15.6
TTFT (s)
Resources & Links
proprietary — Proprietary - API access only
Related Tools
Data sourced from official provider APIs and documentation
Last updated: May 5, 2026
Start building with the right model.
From model selection to production, one platform, no fragmentation.