Claude Haiku 4.5
VS
GPT-5 Mini
Claude Haiku 4.5 vs GPT-5 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.
GPT-5 Mini wins 14 of 22 task areas we tested; Claude Haiku 4.5 takes 8. GPT-5 Mini costs 2.7× less per token ($2.25 vs $6 per 1M).
GPT-5 Mini costs 2.7× less per token ($2.25 vs $6 per 1M).
Task by task
| Task area | Claude Haiku 4.5 | GPT-5 Mini | Winner |
|---|---|---|---|
| Structured Output |
#109
/ 113
Usable
|
#17
/ 113
Excellent
|
GPT-5 Mini |
| Chef / Home Cooking |
#113
/ 126
Needs editing
|
#45
/ 126
Usable
|
GPT-5 Mini |
| Coding |
#94
/ 115
Usable
|
#37
/ 115
Strong
|
GPT-5 Mini |
| Summarization & Meeting Notes |
#27
/ 110
Excellent
|
#81
/ 110
Excellent
|
Claude Haiku 4.5 |
| Presentations & Decks |
#88
/ 110
Strong
|
#39
/ 110
Excellent
|
GPT-5 Mini |
| Training & Education |
#36
/ 110
Excellent
|
#81
/ 110
Strong
|
Claude Haiku 4.5 |
| Legal & HR |
#48
/ 110
Excellent
|
#10
/ 110
Excellent
|
GPT-5 Mini |
| Research & Competitive Analysis |
#59
/ 110
Usable
|
#21
/ 110
Excellent
|
GPT-5 Mini |
| Product & Project Management |
#38
/ 110
Strong
|
#1
/ 110
Excellent
|
GPT-5 Mini |
| Executive Assistant |
#73
/ 112
Strong
|
#108
/ 112
Usable
|
Claude Haiku 4.5 |
| Knowledge & Docs |
#71
/ 110
Usable
|
#41
/ 110
Strong
|
GPT-5 Mini |
| Data & Analytics |
#53
/ 110
Excellent
|
#27
/ 110
Excellent
|
GPT-5 Mini |
| Translation & Localization |
#27
/ 110
Excellent
|
#52
/ 110
Excellent
|
Claude Haiku 4.5 |
| Sales |
#80
/ 110
Usable
|
#102
/ 110
Needs editing
|
Claude Haiku 4.5 |
| Content & Brand |
#89
/ 124
Usable
|
#69
/ 124
Strong
|
GPT-5 Mini |
| RAG, Safety & Grounding |
#97
/ 113
Strong
|
#78
/ 113
Excellent
|
GPT-5 Mini |
| Creative & Comedy | #84 / 110 | #70 / 110 | GPT-5 Mini |
| Investor & Pitch |
#63
/ 66
Needs editing
|
#50
/ 66
Usable
|
GPT-5 Mini |
| Landing Pages |
#61
/ 72
Usable
|
#50
/ 72
Usable
|
GPT-5 Mini |
| AI Strategy |
#91
/ 126
Strong
|
#100
/ 126
Usable
|
Claude Haiku 4.5 |
| Customer Support |
#99
/ 113
Usable
|
#104
/ 113
Usable
|
Claude Haiku 4.5 |
| Frontend & Landing Pages |
#7
/ 109
Usable
|
#12
/ 109
Needs editing
|
Claude Haiku 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 Haiku 4.5 better than GPT-5 Mini?
Across 22 task areas we benchmarked, GPT-5 Mini ranks higher in 14 and Claude Haiku 4.5 in 8.
Which is cheaper, Claude Haiku 4.5 or GPT-5 Mini?
GPT-5 Mini costs 2.7× less per token ($2.25 vs $6 per 1M).
What is Claude Haiku 4.5 better at?
Claude Haiku 4.5 out-ranks GPT-5 Mini at Summarization & Meeting Notes, Training & Education, Executive Assistant.
What is GPT-5 Mini better at?
GPT-5 Mini out-ranks Claude Haiku 4.5 at Structured Output, Chef / Home Cooking, Coding.
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