OpenAIproprietary
Released: 2025-01-31

o3-mini

OpenAI cost-efficient reasoning model optimized for STEM, competitive mathematics, and autonomous coding.

Intelligence Index
89.5/ 100
Calibrated multi-domain composite
Throughput Speed
115tok/sec
Output streaming throughput
Time To First Token
270ms
Initial chunk server latency
LiveBench Contam-Free
80.2%
September 2026 suite score

LiveBench Multi-Domain Evaluation Breakdown

Monthly refreshed contamination-free test sets
Logical Reasoning87.2%
Coding & Repo Repair81.0%
Mathematics (AIME)86.5%
Data Analysis & Tables77.0%
Language & Comprehension74.5%
Instruction Following75.0%

Sourced Benchmark Evaluations (3)

Verified performance across authoritative benchmarks with provenance tracking.
BenchmarkStatusScoreTrustDateSource TypeProvenance
ARC-AGI-2Nearing Saturation68.8%
75
2025-02-28independentSource ↗
SWE-bench VerifiedSaturated63.8%
25
2025-02-09independentSource ↗
IFEvalDeprecated75%
25
2025-02-04independentSource ↗
OpenAI Compatible API ExecutionAPI Ready
from openai import OpenAI

client = OpenAI()
response = client.chat.completions.create(
    model="o3-mini",
    messages=[{"role": "user", "content": "Evaluate multi-step logic problem"}],
    temperature=0.2
)
print(response.choices[0].message.content)

Evaluation & Sourcing Notes

Scores listed for o3-mini represent verified evaluations extracted from official research papers, independent evaluation suites (HELM, LMSYS, OpenCompass, LiveBench), and verified audit reports.

All benchmarks marked as saturated or deprecated reflect historical performance where the benchmark no longer provides active discriminative power.