Claude Haiku 4.5
VS
Grok 4.20 Beta
Claude Haiku 4.5 vs Grok 4.20 Beta: 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.
Grok 4.20 Beta wins 12 of 22 task areas we tested; Claude Haiku 4.5 takes 10. Claude Haiku 4.5 costs 1.3× less per token ($6 vs $8 per 1M).
Claude Haiku 4.5 costs 1.3× less per token ($6 vs $8 per 1M).
Task by task
| Task area | Claude Haiku 4.5 | Grok 4.20 Beta | Winner |
|---|---|---|---|
| Chef / Home Cooking |
#113
/ 126
Needs editing
|
#38
/ 126
Usable
|
Grok 4.20 Beta |
| Summarization & Meeting Notes |
#27
/ 110
Excellent
|
#95
/ 110
Excellent
|
Claude Haiku 4.5 |
| Structured Output |
#109
/ 113
Usable
|
#43
/ 113
Excellent
|
Grok 4.20 Beta |
| Translation & Localization |
#27
/ 110
Excellent
|
#90
/ 110
Strong
|
Claude Haiku 4.5 |
| Legal & HR |
#48
/ 110
Excellent
|
#106
/ 110
Strong
|
Claude Haiku 4.5 |
| AI Strategy |
#91
/ 126
Strong
|
#49
/ 126
Strong
|
Grok 4.20 Beta |
| Training & Education |
#36
/ 110
Excellent
|
#76
/ 110
Strong
|
Claude Haiku 4.5 |
| Data & Analytics |
#53
/ 110
Excellent
|
#91
/ 110
Excellent
|
Claude Haiku 4.5 |
| Sales |
#80
/ 110
Usable
|
#46
/ 110
Usable
|
Grok 4.20 Beta |
| Presentations & Decks |
#88
/ 110
Strong
|
#56
/ 110
Strong
|
Grok 4.20 Beta |
| Customer Support |
#99
/ 113
Usable
|
#72
/ 113
Usable
|
Grok 4.20 Beta |
| Research & Competitive Analysis |
#59
/ 110
Usable
|
#86
/ 110
Needs editing
|
Claude Haiku 4.5 |
| Frontend & Landing Pages |
#7
/ 109
Usable
|
#32
/ 109
Needs editing
|
Claude Haiku 4.5 |
| Investor & Pitch |
#63
/ 66
Needs editing
|
#43
/ 66
Usable
|
Grok 4.20 Beta |
| Coding |
#94
/ 115
Usable
|
#77
/ 115
Strong
|
Grok 4.20 Beta |
| Knowledge & Docs |
#71
/ 110
Usable
|
#86
/ 110
Usable
|
Claude Haiku 4.5 |
| Content & Brand |
#89
/ 124
Usable
|
#99
/ 124
Usable
|
Claude Haiku 4.5 |
| Creative & Comedy | #84 / 110 | #92 / 110 | Claude Haiku 4.5 |
| Product & Project Management |
#38
/ 110
Strong
|
#31
/ 110
Excellent
|
Grok 4.20 Beta |
| Executive Assistant |
#73
/ 112
Strong
|
#69
/ 112
Strong
|
Grok 4.20 Beta |
| RAG, Safety & Grounding |
#97
/ 113
Strong
|
#93
/ 113
Strong
|
Grok 4.20 Beta |
| Landing Pages |
#61
/ 72
Usable
|
#60
/ 72
Usable
|
Grok 4.20 Beta |
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 Grok 4.20 Beta?
Across 22 task areas we benchmarked, Grok 4.20 Beta ranks higher in 12 and Claude Haiku 4.5 in 10.
Which is cheaper, Claude Haiku 4.5 or Grok 4.20 Beta?
Claude Haiku 4.5 costs 1.3× less per token ($6 vs $8 per 1M).
What is Claude Haiku 4.5 better at?
Claude Haiku 4.5 out-ranks Grok 4.20 Beta at Summarization & Meeting Notes, Translation & Localization, Legal & HR.
What is Grok 4.20 Beta better at?
Grok 4.20 Beta out-ranks Claude Haiku 4.5 at Chef / Home Cooking, Structured Output, AI Strategy.
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