Navigate
AI BENCHY
Advertise here

OpenAI: GPT-5.4 Nano vs StepFun: Step 3.7 Flash

GPT-5.4 Nano (medium) leads on average score with 7.5 vs 7.3. GPT-5.4 Nano (medium) has the lower benchmark cost at $0.138 vs $0.454. GPT-5.4 Nano (medium) is faster at 13.24s vs 20.68s, with pass rates of 65.2% vs 68.2%.

Recommended modelGPT-5.4 Nano (medium)It has the best score here (7.5), while costing about 3.3x less than Step 3.7 Flash (low).

Last updated at: 2026-07-17

Metric GPT-5.4 Nano GPT-5.4 Nano medium Release: 2026-03-17 Step 3.7 Flash Step 3.7 Flash low Release: 2026-05-29
Score 7.5 7.3
Rank #51 #64
Reliability 10.0 10.0
Consistency 8.5 8.1
Tests Correct
Attempt pass rate 65.2% 68.2%
Flaky tests 4 5
Total Runs 66 66
Cost per result 1.150 3.782
Total Cost $0.138 $0.454
Input Price $0.200 / 1M $0.200 / 1M
Output Price $1.250 / 1M $1.150 / 1M
Total Input Tokens 82,819 103,833
Output Tokens 7,100 376,581
Reasoning Tokens 90,022 0
Response Time (avg) 13.24s 20.68s
Response Time (max) 94.06s 124.75s
Response Time (total) 291.33s 455.01s

Hamster playing table tennis

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

#51 GPT-5.4 Nano

medium
Cost
$0.007
Time
24.6s
Tokens
4,943 tok

#64 Step 3.7 Flash

low
Invalid SVG
Cost
$0.004
Time
25.3s
Tokens
3,072 tok

Top Models by Score

Score vs Total Cost

Response Time (avg)

Score vs Response Time (avg)

Total Output Tokens

Score vs Total Output Tokens

Category Breakdown

Coding Score Consistency Attempt pass rate Flaky tests Tests Correct Response Time (avg) Input Tokens Output Tokens Reasoning Tokens
GPT-5.4 Nano 6.1 4.7 66.7% 2 19.12s 7,305 516 20,778
Step 3.7 Flash 8.2 7.2 88.9% 1 9.46s 7,437 18,685 0

Quick Compare

Switch Comparison Pair