head-to-head

Anthropic Claude Haiku Latest vs OpenAI: GPT-5.4 Mini

Side-by-side comparison of specs, pricing, benchmark scores, and task rankings. Updated 2026-06-20.

	Anthropic Claude Haiku Latest	OpenAI: GPT-5.4 Mini
Vendor	~anthropic	openai
Quality Score	100	100
Benchmark Score	-	70.8
Input Price	$1.00/M	$0.75/M
Output Price	$5.00/M	$4.50/M
Context Window	200,000	400,000
Max Output	64,000	128,000
Tool Calling	✓	✓
Structured Output	✓	✓
Reasoning Mode	✓	✓
Vision	✓	✓
Audio	-	-
Benchmark Scores
ai_index	-	66.0
ai_index_agentic	-	49.8
ai_index_coding	-	92.5

Who wins by task?

Task	Anthropic Claude Haiku Latest	OpenAI: GPT-5.4 Mini
SQL Generation	129	164
Code Review	124	159
Code Completion	114	132
Code Refactoring	124	158
Bug Fixing	128	170
Unit Test Generation	120	148
Code Documentation	122	140
Regex Writing	118	132
CI/CD Pipelines	116	139
Frontend Component Design	122	141
Data Analysis	124	160
CSV / Spreadsheet Cleanup	125	151
ETL Scripting	120	149
JSON Extraction	122	146
Bulk Data Labeling	120	133
OCR / Document Parsing	127	144
Table Extraction from PDFs	127	144
Long-Document Summarization	125	154
Short-Form Summarization	114	129
Blog Post Writing	117	136

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