Navigate
AI BENCHY
Advertise here

Anthropic: Claude Opus 4.8 vs OpenAI: GPT-5.6 Terra

GPT-5.6 Terra (high) leads on average score with 8.0 vs 7.8. GPT-5.6 Terra (high) has the lower benchmark cost at $1.055 vs $2.077. GPT-5.6 Terra (high) is faster at 11.32s vs 12.74s, with pass rates of 80.3% vs 68.2%.

Recommended modelGPT-5.6 Terra (high)It has the best score here (8.0), while costing about 2.0x less than Claude Opus 4.8 (low).

Last updated at: 2026-07-18

Metric Claude Opus 4.8 Claude Opus 4.8 low Release: 2026-05-28 GPT-5.6 Terra GPT-5.6 Terra high Release: 2026-07-09
Score 7.8 8.0
Rank #41 #34
Reliability 10.0 10.0
Consistency 8.9 9.3
Tests Correct
Attempt pass rate 80.3% 68.2%
Flaky tests 3 2
Total Runs 66 66
Cost per result 12.979 7.533
Total Cost $2.077 $1.055
Input Price $5.000 / 1M $2.500 / 1M
Output Price $25.000 / 1M $15.000 / 1M
Total Input Tokens 156,525 81,047
Output Tokens 43,141 5,055
Reasoning Tokens 8,617 51,736
Response Time (avg) 12.74s 11.32s
Response Time (max) 127.97s 91.49s
Response Time (total) 280.29s 249.14s

Hamster playing table tennis

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

#41 Claude Opus 4.8

low
Cost
$0.031
Time
14.1s
Tokens
1,345 tok

#34 GPT-5.6 Terra

high
Cost
$0.055
Time
30.2s
Tokens
3,748 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 6.6 4.6 77.8% 2 7.58s 10,590 3,637 809
GPT-5.6 Terra 7.6 7.2 77.8% 1 9.14s 7,302 370 6,087

Quick Compare

Switch Comparison Pair