AI BENCHY
Compare
❤️ Made by XCS

Model Name

Anthropic: Claude Sonnet 4.6

Last updated at : Feb 19, 2026

Metric Anthropic: Claude Sonnet 4.6
Rank#13
CompanyAnthropic
Score 5.75
Consistency 9.42
Cost per result 0.9480
Total Cost $0.05688
Tests Correct 6/12
Attempt pass rate 52.8%
Flaky tests 1
Output Tokens 1,659
Reasoning Tokens 0

Category Breakdown

Category Fully passed tests Score Consistency Attempt pass rate Flaky tests Reasoning score Cost
Anti-AI Tricks 0/2 1.00 10.00 0.0% 0 - $0.01092
Data parsing and extraction 2/2 10.00 10.00 100.0% 0 - $0.02854
Domain specific 2/3 7.00 10.00 66.7% 0 - $0.00309
Instructions following 1/2 5.50 10.00 50.0% 0 - $0.00342
Puzzle Solving 1/3 5.00 7.68 44.4% 1 - $0.01092

Compared models

Compare Anthropic: Claude Sonnet 4.6 against...

#12 · OpenAI

OpenAI: gpt-oss-120b

Reasoning (medium)

Score: 5.75

Consistency: 7.19

Attempt pass rate: 63.9%

Flaky tests: 4

Cost per result: 0.0951

Tests Correct: 6/12

Total Cost: $0.00571

Compare

#14 · Qwen

Qwen: Qwen3.5 Plus 2026-02-15

No Reasoning

Score: 5.67

Consistency: 9.99

Attempt pass rate: 50.0%

Flaky tests: 0

Cost per result: 0.0997

Tests Correct: 6/12

Total Cost: $0.00599

Compare

#11 · OpenAI

OpenAI: GPT-5 Nano

Reasoning (medium)

Score: 5.92

Consistency: 6.03

Attempt pass rate: 72.2%

Flaky tests: 6

Cost per result: 0.4675

Tests Correct: 6/12

Total Cost: $0.02806

Compare

Quick Compare

Compare Anthropic: Claude Sonnet 4.6 against...