Spam vs Ham Email

ClassificationDifficulty: EASY11 models comparedTotal cost: $0.000516

Prompt

Classify this email as SPAM or HAM. Reply with only one word.

Subject: Re: Q3 budget review
Body: Hi team, attached is the updated Q3 budget spreadsheet. Please review the marketing line items before Thursday. Best, Sarah

Expected Output

HAM

Model Responses

Claude Opus 4.6

Unrated

2,877ms68 tokens

HAM

GPT-4.1

Unrated

680ms59 tokens

HAM

GPT-4.1 Mini

Unrated

1,895ms59 tokens

HAM

GPT-4.1 Nano

Unrated

1,023ms59 tokens

HAM

o3

Unrated

1,860ms76 tokens

HAM

o3 Mini

Unrated

1,147ms72 tokens

HAM

o4 Mini

Unrated

1,296ms140 tokens

HAM

GPT-4o Mini

Correct

1,422ms59 tokens$0.000009

HAM

Claude Haiku 4.5

Correct

526ms68 tokens$0.000088

HAM

GPT-4o

Correct

5,270ms59 tokens$0.000155

HAM

Claude Sonnet 4.6

Correct

873ms68 tokens$0.000264

HAM

Cost & Performance Comparison

Model	Latency	Tokens	Cost	Verdict
Claude Opus 4.6	2,877ms	68	—	—
GPT-4.1	680ms	59	—	—
GPT-4.1 Mini	1,895ms	59	—	—
GPT-4.1 Nano	1,023ms	59	—	—
o3	1,860ms	76	—	—
o3 Mini	1,147ms	72	—	—
o4 Mini	1,296ms	140	—	—
GPT-4o Mini	1,422ms	59	$0.000009	Correct
Claude Haiku 4.5	526ms	68	$0.000088	Correct
GPT-4o	5,270ms	59	$0.000155	Correct
Claude Sonnet 4.6	873ms	68	$0.000264	Correct