Navigate
AI BENCHY
Advertise here

StepFun: Step 3.5 Flash vs Thinking Machines: Inkling

Inkling (low) leads on average score with 6.1 vs 6.0. Step 3.5 Flash (medium) has the lower benchmark cost at $0.108 vs $0.187. Inkling (low) is faster at 5.15s vs 174.22s, with pass rates of 51.5% vs 54.6%.

Recommended modelInkling (low)It has the best score here (6.1), while responding about 33.8x faster than Step 3.5 Flash (medium).

Last updated at: 2026-07-18

Metric Step 3.5 Flash Step 3.5 Flash medium Release: 2026-02-01 Inkling Inkling low Release: 2026-07-18
Score 6.0 6.1
Rank #126 #119
Reliability 9.2 9.9
Consistency 9.0 8.6
Tests Correct
Attempt pass rate 51.5% 54.6%
Flaky tests 1 4
Total Runs 63 66
Cost per result 0.540 1.866
Total Cost $0.108 $0.187
Input Price $0.100 / 1M $1.000 / 1M
Output Price $0.300 / 1M $4.050 / 1M
Total Input Tokens 65,707 109,884
Output Tokens 108,561 8,579
Reasoning Tokens 293,993 10,343
Response Time (avg) 174.22s 5.15s
Response Time (max) 1597.85s 41.58s
Response Time (total) 2613.32s 113.39s

Hamster playing table tennis

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

#126 Step 3.5 Flash

medium
Cost
$0.008
Time
277.1s
Tokens
23,695 tok

#119 Thinking Machines: Inkling

low
Cost
$0.002
Time
6.1s
Tokens
502 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
Step 3.5 Flash 2.4 5.2 0.0% 0 258.38s 2,211 13,207 22,429
Inkling 5.1 7.2 22.2% 1 8.71s 7,374 355 4,961

Quick Compare

Switch Comparison Pair