head-to-head

Anthropic: Claude Sonnet 5 vs Anthropic: Claude Sonnet Latest

Side-by-side comparison of specs, pricing, benchmark scores, and task rankings. Updated 2026-11-07.

	Anthropic: Claude Sonnet 5	Anthropic: Claude Sonnet Latest
Vendor	anthropic	~anthropic
Quality Score	100	100
Benchmark Score	91.4	-
Input Price	$2.00/M	$2.00/M
Output Price	$10.00/M	$10.00/M
Context Window	1,000,000	1,000,000
Max Output	128,000	128,000
Tool Calling	✓	✓
Structured Output	✓	✓
Reasoning Mode	✓	✓
Vision	✓	✓
Audio	-	-
Benchmark Scores
ai_index	88.0	-
ai_index_agentic	77.0	-
ai_index_coding	100.0	-

The verdict

For a sample job of 1 million input tokens and 250,000 output tokens, both cost $4.50 at current list prices.

Both read up to 1,000,000 tokens at once, roughly 1,500 pages of text.

Across the 20 tasks where they differ, Claude Sonnet 5 scores higher on 20 (SQL Generation, Code Review, Code Completion and others).

Claude Sonnet 5 was released in June 2026 and Claude Sonnet Latest in April 2026.

Pick Claude Sonnet 5 if you need the stronger all-round task scores.

Costs use current list prices per million tokens; page counts assume about 0.75 words per token and 500 words per page.

Task	Anthropic: Claude Sonnet 5	Anthropic: Claude Sonnet Latest
SQL Generation	171	132
Code Review	167	132
Code Completion	119	117
Code Refactoring	164	136
Bug Fixing	180	136
Unit Test Generation	154	124
Code Documentation	141	129
Regex Writing	135	117
CI/CD Pipelines	144	120
Frontend Component Design	146	122
Data Analysis	171	124
CSV / Spreadsheet Cleanup	153	132
ETL Scripting	154	128
JSON Extraction	138	121
Bulk Data Labeling	124	118
OCR / Document Parsing	146	131
Table Extraction from PDFs	146	131
Long-Document Summarization	159	136
Short-Form Summarization	122	113
Blog Post Writing	140	120

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