Codesota · Models · o4-miniOpenAI15 results · 15 benchmarks
Model card

o4-mini.

OpenAIapi3 current SOTA
§ 01 · Benchmarks

Every benchmark o4-mini has a recorded score for.

#BenchmarkArea · TaskMetricValueRankDateSource
01AIME 2025Reasoning · Mathematical Reasoningaccuracy92.7%#1/5—source ↗
02HumanEvalComputer Code · Code Generationpass@197.3%#1/42—source ↗
03MBPPComputer Code · Code Generationpass@194.9%#1/19—source ↗
04AIME 2024Reasoning · Mathematical Reasoningaccuracy93.4%#2/8—source ↗
05ARC-AGI-1Reasoning · Logical Reasoningaccuracy79.0%#3/5—source ↗
06ARC-AGI-2Reasoning · Logical Reasoningaccuracy3.0%#3/3—source ↗
07GSM8KReasoning · Mathematical Reasoningaccuracy99.0%#3/32—source ↗
08LiveCodeBench ProComputer Code · Code Generationelo2092.00#3/9—source ↗
09ARC-ChallengeReasoning · Commonsense Reasoningaccuracy97.3%#4/10—source ↗
10MATHReasoning · Mathematical Reasoningaccuracy97.5%#5/34—source ↗
11LiveCodeBenchComputer Code · Code Generationpass@172.8%#7/302024-03-12source ↗
12GPQAReasoning · Multi-step Reasoningaccuracy77.6%#10/33—source ↗
13MMLUReasoning · Commonsense Reasoningaccuracy90.0%#15/412025-04-16source ↗
14SWE-Bench VerifiedComputer Code · Code Generationresolve-rate68.1%#22/39—source ↗
15SWE-bench VerifiedAgentic AI · SWE-benchresolve-rate68.1%#45/81—source ↗
Rank column shows this model’s position vs all other models scored on the same benchmark + metric (competitors after the slash). #1 in red means current SOTA. Sorted by rank, then newest result.
§ 02 · Strengths by area

Where o4-mini actually performs.

Computer Code
5
benchmarks
avg rank #6.8 · 2 SOTA
Reasoning
9
benchmarks
avg rank #5.1 · 1 SOTA
Agentic AI
1
benchmark
avg rank #45.0
§ 03 · Papers

1 paper with results for o4-mini.

  1. 2024-03-12· Computer Code· 1 result

    LiveCodeBench: Holistic and Contamination Free Evaluation of Large Language Models for Code

§ 04 · Related models

Other OpenAI models scored on Codesota.

GPT-4o
Undisclosed params · 35 results · 9 SOTA
o3
16 results · 5 SOTA
o3 (high)
2 results · 1 SOTA
o4-mini (high)
1 result · 1 SOTA
o1
11 results
GPT-5
8 results
o1-preview
Undisclosed params · 8 results
GPT-4.1
7 results
§ 05 · Sources & freshness

Where these numbers come from.

openai-simple-evals
6
results
openai-system-card
2
results
arcprize-leaderboard
2
results
official-model-card
1
result
livecodebench-pro-official
1
result
official-leaderboard
1
result
swebench-leaderboard
1
result
editorial
1
result
11 of 15 rows marked verified. · first result 2024-03-12, latest 2025-04-16.