#65 GPT-5.4 Mini
medium- Cost
- $0.056
- Time
- 95.5s
- Tokens
- 12,464 tok
OpenAI
Release: 2026-03-17
Tested on: 2026-05-22 00:21
openai/gpt-5.4-mini::medium
~400B total (~17B active)MoEClosedEstimated
(medium)
(none)
Summary
GPT-5.4 Mini scores 7.1 on AI BENCHY and ranks #65. It has 10.0 reliability, a 68.3% pass rate, $0.487 total cost, and 22.14s average response time.
Model facts
Researched on 2026-08-12
- Parameters
- ~400B total (~17B active)
- Architecture
- MoE
- Availability
- Closed
- License
- -
Best estimate from public evidence; the vendor did not disclose every value. Vendor does not disclose the architecture or count.
7.1
Consistency
7.6
10.0
$0.487
Total Output Tokens
102,892
Total Input Tokens
0
Input Price
$0.750 / 1M
Output Price
$4.500 / 1M
Flaky tests
6
Flaky tests had mixed outcomes across runs (at least one pass and one fail).
Hamster playing table tennis
Prompt: Create a detailed SVG illustration of a hamster playing table tennis.
Run history
| Tested on | Score | Reliability | Tests Correct | Total Cost | Compare |
|---|---|---|---|---|---|
| 2026-08-14 02:19 Re-test | 7.5 | 10.0 | $0.786 | Compare | |
| 2026-07-16 22:38 New test added | 7.5 | 10.0 | $0.756 | Compare | |
| 2026-06-04 13:47 New test added | 7.5 | 10.0 | $0.526 | Compare | |
| 2026-05-22 00:21 Suite changed | 7.1 | 10.0 | $0.487 | Current run | |
| 2026-04-11 01:44 First recorded run | 7.3 | N/A | $0.299 | Compare |
This run used a different benchmark suite. Keep suite changes in mind when reading historical movement.
Run comparison
| Run | Benchmark coverage | Score | Consistency | Reliability | Tests Correct | Flaky tests | Total Output Tokens | Total Input Tokens | Total Cost | Response Time (avg) |
|---|---|---|---|---|---|---|---|---|---|---|
| 2026-05-22 00:21 · Suite changed | 60/60 attempts | 7.1 | 7.6 | 10.0 | 10/20 | 6 | 102,892 | 0 | $0.487 | 22.14s |
| 2026-04-11 01:44 · First recorded run | 54/54 attempts | 7.3 | 7.4 | N/A | 9/18 | 6 | 61,698 | 0 | $0.299 | 15.22s |
| Difference | — | -0.2 | +0.2 | +1 | 0 | +41194 | 0 | +$0.189 | +6920ms |
Benchmark coverage differs: 60/60 attempts (Target: 3 repeats per test) versus 54/54 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.
Top Models by Score
Score vs Total Cost
Response Time (avg)
Score vs Response Time (avg)
Total Output Tokens
Score vs Total Output Tokens
Quick Compare
GPT-5.4 MinimediumvsMercury 2.5 PreviewmediumGPT-5.4 MinimediumvsGrok Build 0.1mediumGPT-5.4 MinimediumvsQwen3.5-27BmediumGPT-5.4 MinimediumvsGemini 3.8 FlashlowGPT-5.4 MinimediumvsQwen3.5 Plus 2026-02-15mediumGPT-5.4 MinimediumvsGemini 3.1 Flash Lite PreviewmediumGPT-5.4 MinimediumvsGPT 5.3 ChatnoneGPT-5.4 MinimediumvsQwen3.7 MaxnoneGPT-5.4 MinimediumvsMiniMax M3mediumGPT-5.4 MinimediumvsKimi K2.7 Codemedium
Category Breakdown
| Category | Score | Consistency | Tests Correct |
|---|---|---|---|
| Anti-AI Tricks | 8.6 | 7.9 | |
| Coding | 7.5 | 6.0 | |
| Combined | 10.0 | 10.0 | |
| Data parsing and extraction | 10.0 | 10.0 | |
| Domain specific | 4.1 | 4.4 | |
| General Intelligence | 4.5 | 10.0 | |
| Instructions following | 7.4 | 6.7 | |
| Puzzle Solving | 7.8 | 10.0 | |
| Tool Calling | 4.7 | 1.6 | |
| Trivia | 3.0 | 10.0 |