Navigate
AI BENCHY
Advertise here

Anthropic: Claude Opus 4.8 vs OpenAI: GPT-5.4

Claude Opus 4.8 (medium) leads on average score with 8.8 vs 8.5. GPT-5.4 (medium) has the lower benchmark cost at $1.533 vs $1.931. Claude Opus 4.8 (medium) is faster at 12.49s vs 23.10s, with pass rates of 84.9% vs 77.3%.

Recommended modelClaude Opus 4.8 (medium)It has the best score here (8.8), while responding about 1.9x faster than GPT-5.4 (medium).

Last updated at: 2026-07-22

Metric Claude Opus 4.8 Claude Opus 4.8 medium Release: 2026-05-28 GPT-5.4 GPT-5.4 medium Release: 2026-03-05
Score 8.8 8.5
Rank #17 #21
Reliability 10.0 10.0
Consistency 9.6 8.6
Tests Correct
Attempt pass rate 84.9% 77.3%
Flaky tests 1 4
Total Runs 66 66
Cost per result 10.724 10.220
Total Cost $1.931 $1.533
Input Price $5.000 / 1M $2.500 / 1M
Output Price $25.000 / 1M $15.000 / 1M
Total Input Tokens 138,451 81,127
Output Tokens 40,766 6,155
Reasoning Tokens 9,075 82,515
Response Time (avg) 12.49s 23.10s
Response Time (max) 70.54s 100.41s
Response Time (total) 274.72s 508.26s

Hamster playing table tennis

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

#17 Claude Opus 4.8

medium
Cost
$0.057
Time
23.1s
Tokens
2,412 tok

#21 GPT-5.4

medium
Cost
$0.214
Time
199.6s
Tokens
14,349 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 10.0 10.0 100.0% 0 15.33s 10,590 9,945 1,381
GPT-5.4 8.8 7.8 88.9% 1 44.36s 7,305 433 24,216

Quick Compare

Switch Comparison Pair