Claude Sonnet 4.5
VS
GPT-5.4 Mini
Claude Sonnet 4.5 vs GPT-5.4 Mini: which wins at real work?
22 task areas · same graded test runs · rank comparison only, so 0–100 and Elo collections never mix raw scores.
Claude Sonnet 4.5 and GPT-5.4 Mini split the 22 task areas we tested 11–11. GPT-5.4 Mini costs 3.4× less per token ($5.25 vs $18 per 1M).
GPT-5.4 Mini costs 3.4× less per token ($5.25 vs $18 per 1M).
Task by task
| Task area | Claude Sonnet 4.5 | GPT-5.4 Mini | Winner |
|---|---|---|---|
| Structured Output |
#80
/ 113
Strong
|
#1
/ 113
Excellent
|
GPT-5.4 Mini |
| Executive Assistant |
#31
/ 112
Strong
|
#87
/ 112
Usable
|
Claude Sonnet 4.5 |
| Product & Project Management |
#71
/ 110
Strong
|
#19
/ 110
Excellent
|
GPT-5.4 Mini |
| Legal & HR |
#69
/ 110
Strong
|
#26
/ 110
Excellent
|
GPT-5.4 Mini |
| Coding |
#62
/ 115
Strong
|
#24
/ 115
Strong
|
GPT-5.4 Mini |
| Data & Analytics |
#33
/ 110
Excellent
|
#65
/ 110
Excellent
|
Claude Sonnet 4.5 |
| Sales |
#10
/ 110
Strong
|
#40
/ 110
Usable
|
Claude Sonnet 4.5 |
| Research & Competitive Analysis |
#39
/ 110
Strong
|
#11
/ 110
Excellent
|
GPT-5.4 Mini |
| Training & Education |
#51
/ 110
Excellent
|
#79
/ 110
Strong
|
Claude Sonnet 4.5 |
| Presentations & Decks |
#43
/ 110
Excellent
|
#16
/ 110
Excellent
|
GPT-5.4 Mini |
| Knowledge & Docs |
#28
/ 110
Strong
|
#3
/ 110
Excellent
|
GPT-5.4 Mini |
| Customer Support |
#43
/ 113
Strong
|
#21
/ 113
Strong
|
GPT-5.4 Mini |
| AI Strategy |
#67
/ 126
Strong
|
#48
/ 126
Strong
|
GPT-5.4 Mini |
| Summarization & Meeting Notes |
#2
/ 110
Excellent
|
#19
/ 110
Excellent
|
Claude Sonnet 4.5 |
| Landing Pages |
#30
/ 72
Strong
|
#45
/ 72
Usable
|
Claude Sonnet 4.5 |
| Content & Brand |
#74
/ 124
Usable
|
#84
/ 124
Usable
|
Claude Sonnet 4.5 |
| Creative & Comedy | #66 / 110 | #56 / 110 | GPT-5.4 Mini |
| Investor & Pitch |
#42
/ 66
Usable
|
#52
/ 66
Usable
|
Claude Sonnet 4.5 |
| Frontend & Landing Pages |
#29
/ 109
Needs editing
|
#38
/ 109
Needs editing
|
Claude Sonnet 4.5 |
| Translation & Localization |
#25
/ 110
Excellent
|
#21
/ 110
Excellent
|
GPT-5.4 Mini |
| Chef / Home Cooking |
#80
/ 126
Usable
|
#83
/ 126
Usable
|
Claude Sonnet 4.5 |
| RAG, Safety & Grounding |
#39
/ 113
Excellent
|
#42
/ 113
Excellent
|
Claude Sonnet 4.5 |
Rank = position among every model config we tested in that task area (lower is better). Sorted by biggest gap first.
Frequently asked
Is Claude Sonnet 4.5 better than GPT-5.4 Mini?
They tie across the 22 task areas we benchmarked, at 11 apiece.
Which is cheaper, Claude Sonnet 4.5 or GPT-5.4 Mini?
GPT-5.4 Mini costs 3.4× less per token ($5.25 vs $18 per 1M).
What is Claude Sonnet 4.5 better at?
Claude Sonnet 4.5 out-ranks GPT-5.4 Mini at Executive Assistant, Data & Analytics, Sales.
What is GPT-5.4 Mini better at?
GPT-5.4 Mini out-ranks Claude Sonnet 4.5 at Structured Output, Product & Project Management, Legal & HR.
More comparisons
This page is Spring Prompt, running in public
We just did this for every model. Do it for your prompt.
The rankings above come from running real tasks through real models and scoring every output. Spring Prompt is that same engine — pointed at your prompt, your test cases, and your definition of good.
- Generate test cases from your prompt — no eval set required to start.
- Compare models side by side with quality, cost and latency in one matrix.
- Optimise the winner until the scores say it's ready to ship.
Prompt × model results
12 test cases · 3 evals