Navigate
AI BENCHY
AD
Track all your projects in one dashboard. Get 📊stats, 🔥heatmaps and 👀recordings in one self-hosted dashboard.
uxwizz.com

AI BENCHY Compare

Anthropic: Claude Opus 4.8 vs StepFun: Step 3.7 Flash

Summary

Claude Opus 4.8 leads on average score with 7.2 vs 7.1. Claude Opus 4.8 has the lower benchmark cost at $0.539 vs $1.148. Claude Opus 4.8 is faster at 3.48s vs 64.46s, with pass rates of 61.9% vs 63.5%.

Recommended modelClaude Opus 4.8It has the best score here (7.2), while costing about 2.1x less than Step 3.7 Flash (high).

Last updated at: 2026-07-14

Metric Claude Opus 4.8 Claude Opus 4.8 none Release: 2026-05-28 Step 3.7 Flash Step 3.7 Flash high Release: 2026-05-29
Score 7.2 7.1
Rank #73 #79
Reliability 10.0 10.0
Consistency 9.2 8.2
Tests Correct
Attempt pass rate 61.9% 63.5%
Flaky tests 2 4
Total Runs 63 63
Cost per result 4.485 10.434
Total Cost $0.539 $1.148
Input Price $5.000 / 1M $0.200 / 1M
Output Price $25.000 / 1M $1.150 / 1M
Total Input Tokens 67,104 38,391
Output Tokens 8,107 991,355
Reasoning Tokens 0 0
Response Time (avg) 3.48s 64.46s
Response Time (max) 17.73s 364.99s
Response Time (total) 73.00s 1353.57s

Generation showcase

Hamster playing table tennis

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

#73 Claude Opus 4.8

none
Cost
$0.053
Time
22.0s
Tokens
2,253 tok

#79 Step 3.7 Flash

high
Cost
$0.007
Time
63.6s
Tokens
6,030 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
Claude Opus 4.8 5.5 10.0 33.3% 0 3.29s 10,590 1,332 0
Step 3.7 Flash 4.0 6.0 22.2% 1 206.21s 6,057 327,340 0

Quick Compare

Switch Comparison Pair