AI BENCHY
Advertise here
#97

GPT-4o-mini

OpenAI Release: 2024-07-18 Tested on: 2026-04-11 01:19 openai/gpt-4o-mini::none
~8BDenseClosedEstimated

Summary

GPT-4o-mini scores 4.9 on AI BENCHY and ranks #97. It has N/A reliability, a 22.2% pass rate, $0.005 total cost, and 2.00s average response time.

What makes GPT-4o-mini unique: Its total benchmark cost is unusually low for its score range. It is notably fast compared with similar models.

Model facts

Researched on 2026-08-12

Estimated
Parameters
~8B
Architecture
Dense
Availability
Closed
License
-

Best estimate from public evidence; the vendor did not disclose every value. OpenAI does not disclose the parameter count.

Parameter source Availability source

Score

4.9

Consistency

9.9

Reliability

N/A

Total Cost (Current Price)

$0.005

Total Output Tokens

1,947

Total Input Tokens

0

Input Price

$0.150 / 1M

Output Price

$0.600 / 1M

Tests Correct

Wrong Tests: 14

Attempt pass rate: 22.2%

Flaky tests

0

Flaky tests had mixed outcomes across runs (at least one pass and one fail).

Response Time (avg)

2.00s

Response Time (max): 7.58s

Response Time (total): 21.99s

Wrong answer: 13 Did not follow instructions: 1

Hamster playing table tennis

Prompt: Create a detailed SVG illustration of a hamster playing table tennis.

#97 GPT-4o-mini

none
Cost
$0.001
Time
6.6s
Tokens
742 tok

Run history

Tested on Score Reliability Tests Correct Total Cost Compare
2026-08-14 01:32 Re-test 5.0 10.0 $0.010 Compare
2026-07-16 21:13 New test added 5.0 10.0 $0.010 Compare
2026-06-04 13:04 New test added 4.8 10.0 $0.006 Compare
2026-05-21 23:41 Suite changed 4.9 10.0 $0.006 Compare
2026-04-11 01:19 First recorded run 4.9 N/A $0.005 Current run

Run comparison

RunBenchmark coverageScoreConsistencyReliabilityTests CorrectFlaky testsTotal Output TokensTotal Input TokensTotal CostResponse Time (avg)
2026-04-11 01:19 · First recorded run54/54 attempts4.99.9N/A4/1801,9470$0.0052.00s
2026-05-21 23:41 · Suite changed60/60 attempts4.99.910.05/2001,9770$0.0061.87s
Difference—0.00.0-10-300-$0.001+127ms

Benchmark coverage differs: 54/54 attempts (Target: 3 repeats per test) versus 60/60 attempts (Target: 3 repeats per test). Totals and repeat-sensitive metrics are not directly comparable.

These two runs used different benchmark suites, so the deltas reflect both model changes and suite changes.

Charts

Choose the first model, then click a second model to open a side-by-side page.

Total Output Tokens

Score vs Total Output Tokens

Quick Compare

GPT-4o-mininonevsDeepSeek V3.2noneGPT-4o-mininonevsLing-2.6-flashnoneGPT-4o-mininonevsLaguna S 2.1lowFree AvailableGPT-4o-mininonevsNemotron 3 SupernoneFree AvailableGPT-4o-mininonevsMiniMax M2.7mediumGPT-4o-mininonevsRing-2.6-1TnoneGPT-4o-mininonevsNorth Mini CodenoneFree AvailableGPT-4o-mininonevsNemotron 3.5 LightninglowFree AvailableGPT-4o-mininonevsQwen3.5-9BnoneGPT-4o-mininonevsGPT-5.4 Nanonone

Category Breakdown

Category Score Consistency Tests Correct
Anti-AI Tricks 4.8 10.0
Coding 3.0 8.7
Combined 3.0 10.0
Data parsing and extraction 10.0 10.0
Domain specific 3.0 10.0
General Intelligence 4.0 10.0
Instructions following 4.8 10.0
Puzzle Solving 3.7 10.0
Tool Calling 10.0 10.0

Compared models