Navigate
AI BENCHY
Advertise here

Claude Opus 4.8 (medium) vs GPT-5.6 Sol (high)

GPT-5.6 Sol (high) leads on average score with 9.4 vs 8.8. GPT-5.6 Sol (high) has the lower benchmark cost at $1.234 vs $1.931. GPT-5.6 Sol (high) is faster at 11.73s vs 12.49s, with pass rates of 84.9% vs 89.4%.

Last updated at: 2026-08-07

Rank
#19
Total Output Tokens
49,841
Response Time (avg)
12.49s
Total Cost
$1.931
Rank
#7
Total Output Tokens
27,899
Response Time (avg)
11.73s
Total Cost
$1.234
Recommended model GPT-5.6 Sol (high)

It has the best score here (9.4), while costing about 1.6x less than Claude Opus 4.8 (medium).

Detailed comparison

Metric Claude Opus 4.8 Claude Opus 4.8 medium Release: 2026-05-28 GPT-5.6 Sol GPT-5.6 Sol high Release: 2026-07-09
Score 8.8 9.4
Rank #19 #7
Reliability 10.0 10.0
Consistency 9.6 8.9
Benchmark coverage 22/22 tests · 66/66 attempts 22/22 tests · 66/66 attempts
Tests Correct
Attempt pass rate 84.9% 89.4%
Flaky tests 1 3
Total Runs 66 66
Cost per result 10.724 6.852
Total Cost $1.931 $1.234
Input Price $5.000 / 1M $5.000 / 1M
Output Price $25.000 / 1M $30.000 / 1M
Total Input Tokens 138,451 79,249
Output Tokens 40,766 4,855
Reasoning Tokens 9,075 23,044
Response Time (avg) 12.49s 11.73s
Response Time (max) 70.54s 54.79s
Response Time (total) 274.72s 257.99s

Model generation showcase

Hamster playing table tennis

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

#19 Claude Opus 4.8

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

#7 GPT-5.6 Sol

high
Cost
$0.151
Time
201.9s
Tokens
5,100 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-5.6 Sol 10.0 10.0 100.0% 0 12.52s 7,302 404 5,656

Quick Compare

Switch Comparison Pair