picksbymodel Get API access →
head-to-head

OpenAI: GPT-5.3-Codex vs Anthropic: Claude Sonnet 4.6

Side-by-side comparison of specs, pricing, benchmark scores, and task rankings. Updated 2026-05-12.

OpenAI: GPT-5.3-Codex Anthropic: Claude Sonnet 4.6
Vendoropenaianthropic
Quality Score100100
Benchmark Score-80.0
Input Price$1.75/M$3.00/M
Output Price$14.00/M$15.00/M
Context Window400,0001,000,000
Max Output128,000128,000
Tool Calling✓✓
Structured Output✓✓
Reasoning Mode✓✓
Vision✓✓
Audio--
Benchmark Scores
ai_index-85.3
ai_index_agentic-100.0
ai_index_coding-84.1
aider_polyglot-61.3

Who wins by task?

TaskOpenAI: GPT-5.3-CodexAnthropic: Claude Sonnet 4.6
SQL Generation 132 181
Code Review 132 177
Code Completion 116 118
Code Refactoring 136 172
Bug Fixing 136 194
Unit Test Generation 124 163
Code Documentation 128 144
Regex Writing 116 139
CI/CD Pipelines 120 152
Frontend Component Design 122 153
Data Analysis 124 184
CSV / Spreadsheet Cleanup 132 158
ETL Scripting 128 162
JSON Extraction 120 141
Bulk Data Labeling 117 123
OCR / Document Parsing 131 150
Table Extraction from PDFs 131 150
Long-Document Summarization 136 166
Short-Form Summarization 112 123
Blog Post Writing 120 145

Scores reflect capability match + benchmark data + pricing for each task. Methodology →

Related comparisons

Google: Gemini 3.1 Flash Lite vs OpenAI: GPT-5.3-Codex Google: Gemini 3.1 Flash Lite vs Anthropic: Claude Sonnet 4.6 xAI: Grok 4.3 vs OpenAI: GPT-5.3-Codex xAI: Grok 4.3 vs Anthropic: Claude Sonnet 4.6 Mistral: Mistral Medium 3.5 vs OpenAI: GPT-5.3-Codex Mistral: Mistral Medium 3.5 vs Anthropic: Claude Sonnet 4.6 NVIDIA: Nemotron 3 Nano Omni (free) vs OpenAI: GPT-5.3-Codex NVIDIA: Nemotron 3 Nano Omni (free) vs Anthropic: Claude Sonnet 4.6