Claude Haiku 4.5 vs Claude Sonnet 5: 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 5 wins 16 of 22 task areas we tested; Claude Haiku 4.5 takes 6. Claude Haiku 4.5 costs 2.0× less per token ($6 vs $12 per 1M).
Claude Haiku 4.5 costs 2.0× less per token ($6 vs $12 per 1M).
Task by task
| Task area | Claude Haiku 4.5 | Claude Sonnet 5 | Winner |
|---|---|---|---|
| AI Strategy |
#91
/ 126
Strong
|
#2
/ 126
Strong
|
Claude Sonnet 5 |
| Coding |
#94
/ 115
Usable
|
#6
/ 115
Excellent
|
Claude Sonnet 5 |
| Chef / Home Cooking |
#113
/ 126
Needs editing
|
#28
/ 126
Usable
|
Claude Sonnet 5 |
| Presentations & Decks |
#88
/ 110
Strong
|
#22
/ 110
Excellent
|
Claude Sonnet 5 |
| Sales |
#80
/ 110
Usable
|
#16
/ 110
Strong
|
Claude Sonnet 5 |
| Knowledge & Docs |
#71
/ 110
Usable
|
#18
/ 110
Strong
|
Claude Sonnet 5 |
| Frontend & Landing Pages |
#7
/ 109
Usable
|
#57
/ 109
Needs editing
|
Claude Haiku 4.5 |
| Landing Pages |
#61
/ 72
Usable
|
#12
/ 72
Strong
|
Claude Sonnet 5 |
| Research & Competitive Analysis |
#59
/ 110
Usable
|
#15
/ 110
Excellent
|
Claude Sonnet 5 |
| Customer Support |
#99
/ 113
Usable
|
#57
/ 113
Strong
|
Claude Sonnet 5 |
| Structured Output |
#109
/ 113
Usable
|
#69
/ 113
Strong
|
Claude Sonnet 5 |
| Data & Analytics |
#53
/ 110
Excellent
|
#16
/ 110
Excellent
|
Claude Sonnet 5 |
| Summarization & Meeting Notes |
#27
/ 110
Excellent
|
#58
/ 110
Excellent
|
Claude Haiku 4.5 |
| Investor & Pitch |
#63
/ 66
Needs editing
|
#39
/ 66
Usable
|
Claude Sonnet 5 |
| Product & Project Management |
#38
/ 110
Strong
|
#61
/ 110
Strong
|
Claude Haiku 4.5 |
| Content & Brand |
#89
/ 124
Usable
|
#67
/ 124
Strong
|
Claude Sonnet 5 |
| Training & Education |
#36
/ 110
Excellent
|
#52
/ 110
Excellent
|
Claude Haiku 4.5 |
| Translation & Localization |
#27
/ 110
Excellent
|
#41
/ 110
Excellent
|
Claude Haiku 4.5 |
| RAG, Safety & Grounding |
#97
/ 113
Strong
|
#86
/ 113
Excellent
|
Claude Sonnet 5 |
| Legal & HR |
#48
/ 110
Excellent
|
#57
/ 110
Excellent
|
Claude Haiku 4.5 |
| Creative & Comedy | #84 / 110 | #79 / 110 | Claude Sonnet 5 |
| Executive Assistant |
#73
/ 112
Strong
|
#72
/ 112
Strong
|
Claude Sonnet 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 Claude Sonnet 5?
Across 22 task areas we benchmarked, Claude Sonnet 5 ranks higher in 16 and Claude Haiku 4.5 in 6.
Which is cheaper, Claude Haiku 4.5 or Claude Sonnet 5?
Claude Haiku 4.5 costs 2.0× less per token ($6 vs $12 per 1M).
What is Claude Haiku 4.5 better at?
Claude Haiku 4.5 out-ranks Claude Sonnet 5 at Frontend & Landing Pages, Summarization & Meeting Notes, Product & Project Management.
What is Claude Sonnet 5 better at?
Claude Sonnet 5 out-ranks Claude Haiku 4.5 at AI Strategy, Coding, Chef / Home Cooking.
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