Navigate
AI BENCHY
Compare Charts
❤️ Made by XCS
Your ad here

AI BENCHY Compare

Anthropic: Claude Sonnet 4.6 vs OpenAI: GPT-5 Nano

Model Name:

Last updated at : 2026-02-27 15:16

Summary

Metric Anthropic: Claude Sonnet 4.6 medium Release: Unknown release date OpenAI: GPT-5 Nano medium Release: Unknown release date
Rank #8 #23
Score 7.43 5.86
Consistency 9.40 6.60
Cost per result 8.105 0.519
Total Cost $0.811 $0.037
Tests Correct
Wrong Tests 4 7
Attempt pass rate 73.8% 69.1%
Flaky tests 1 6
Output Tokens 29,098 3,700
Reasoning Tokens 20,435 85,184

Category Breakdown

Anti-AI Tricks Score Consistency Attempt pass rate Flaky tests Tests Correct Output Tokens Reasoning Tokens
Anthropic: Claude Sonnet 4.6 7.00 10.00 66.7% 0 1,031 1,093
OpenAI: GPT-5 Nano 7.00 9.99 66.7% 0 1,107 19,968
Data parsing and extraction Score Consistency Attempt pass rate Flaky tests Tests Correct Output Tokens Reasoning Tokens
Anthropic: Claude Sonnet 4.6 10.00 10.00 100.0% 0 727 907
OpenAI: GPT-5 Nano 5.50 5.81 83.3% 1 426 8,576
Domain specific Score Consistency Attempt pass rate Flaky tests Tests Correct Output Tokens Reasoning Tokens
Anthropic: Claude Sonnet 4.6 1.00 7.21 11.1% 1 25,790 16,919
OpenAI: GPT-5 Nano 4.00 4.41 55.6% 2 195 33,600
Instructions following Score Consistency Attempt pass rate Flaky tests Tests Correct Output Tokens Reasoning Tokens
Anthropic: Claude Sonnet 4.6 10.00 10.00 100.0% 0 316 523
OpenAI: GPT-5 Nano 7.00 6.41 83.3% 1 360 4,032
Puzzle Solving Score Consistency Attempt pass rate Flaky tests Tests Correct Output Tokens Reasoning Tokens
Anthropic: Claude Sonnet 4.6 10.00 10.00 100.0% 0 579 642
OpenAI: GPT-5 Nano 4.67 4.90 55.6% 2 1,054 12,032
Tool Calling Score Consistency Attempt pass rate Flaky tests Tests Correct Output Tokens Reasoning Tokens
Anthropic: Claude Sonnet 4.6 10.00 10.00 100.0% 0 655 351
OpenAI: GPT-5 Nano 10.00 10.00 100.0% 0 558 6,976

Switch Comparison Pair

Compare Anthropic: Claude Sonnet 4.6 against...

Compare OpenAI: GPT-5 Nano against...