Claude Haiku 4.5
VS
Claude Sonnet 4.6
Claude Haiku 4.5 vs Claude Sonnet 4.6: 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 4.6 wins 19 of 22 task areas we tested; Claude Haiku 4.5 takes 3. Claude Haiku 4.5 costs 3.0× less per token ($6 vs $18 per 1M).
Claude Haiku 4.5 costs 3.0× less per token ($6 vs $18 per 1M).
Task by task
| Task area | Claude Haiku 4.5 | Claude Sonnet 4.6 | Winner |
|---|---|---|---|
| Chef / Home Cooking |
#113
/ 126
Needs editing
|
#20
/ 126
Usable
|
Claude Sonnet 4.6 |
| AI Strategy |
#91
/ 126
Strong
|
#3
/ 126
Strong
|
Claude Sonnet 4.6 |
| Executive Assistant |
#73
/ 112
Strong
|
#1
/ 112
Strong
|
Claude Sonnet 4.6 |
| Sales |
#80
/ 110
Usable
|
#13
/ 110
Strong
|
Claude Sonnet 4.6 |
| Coding |
#94
/ 115
Usable
|
#29
/ 115
Strong
|
Claude Sonnet 4.6 |
| Creative & Comedy | #84 / 110 | #19 / 110 | Claude Sonnet 4.6 |
| Investor & Pitch |
#63
/ 66
Needs editing
|
#4
/ 66
Strong
|
Claude Sonnet 4.6 |
| Content & Brand |
#89
/ 124
Usable
|
#31
/ 124
Strong
|
Claude Sonnet 4.6 |
| Landing Pages |
#61
/ 72
Usable
|
#5
/ 72
Strong
|
Claude Sonnet 4.6 |
| Presentations & Decks |
#88
/ 110
Strong
|
#35
/ 110
Excellent
|
Claude Sonnet 4.6 |
| Data & Analytics |
#53
/ 110
Excellent
|
#6
/ 110
Excellent
|
Claude Sonnet 4.6 |
| Legal & HR |
#48
/ 110
Excellent
|
#3
/ 110
Excellent
|
Claude Sonnet 4.6 |
| Customer Support |
#99
/ 113
Usable
|
#62
/ 113
Strong
|
Claude Sonnet 4.6 |
| Knowledge & Docs |
#71
/ 110
Usable
|
#35
/ 110
Strong
|
Claude Sonnet 4.6 |
| Training & Education |
#36
/ 110
Excellent
|
#9
/ 110
Excellent
|
Claude Sonnet 4.6 |
| RAG, Safety & Grounding |
#97
/ 113
Strong
|
#73
/ 113
Excellent
|
Claude Sonnet 4.6 |
| Research & Competitive Analysis |
#59
/ 110
Usable
|
#35
/ 110
Strong
|
Claude Sonnet 4.6 |
| Frontend & Landing Pages |
#7
/ 109
Usable
|
#23
/ 109
Needs editing
|
Claude Haiku 4.5 |
| Translation & Localization |
#27
/ 110
Excellent
|
#13
/ 110
Excellent
|
Claude Sonnet 4.6 |
| Product & Project Management |
#38
/ 110
Strong
|
#48
/ 110
Strong
|
Claude Haiku 4.5 |
| Summarization & Meeting Notes |
#27
/ 110
Excellent
|
#37
/ 110
Excellent
|
Claude Haiku 4.5 |
| Structured Output |
#109
/ 113
Usable
|
#105
/ 113
Usable
|
Claude Sonnet 4.6 |
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 4.6?
Across 22 task areas we benchmarked, Claude Sonnet 4.6 ranks higher in 19 and Claude Haiku 4.5 in 3.
Which is cheaper, Claude Haiku 4.5 or Claude Sonnet 4.6?
Claude Haiku 4.5 costs 3.0× less per token ($6 vs $18 per 1M).
What is Claude Haiku 4.5 better at?
Claude Haiku 4.5 out-ranks Claude Sonnet 4.6 at Frontend & Landing Pages, Product & Project Management, Summarization & Meeting Notes.
What is Claude Sonnet 4.6 better at?
Claude Sonnet 4.6 out-ranks Claude Haiku 4.5 at Chef / Home Cooking, AI Strategy, Executive Assistant.
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