Grok 4.20 Beta
VS
MiniMax M3
Grok 4.20 Beta vs MiniMax M3: 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.
MiniMax M3 wins 19 of 22 task areas we tested; Grok 4.20 Beta takes 3. MiniMax M3 costs 5.3× less per token ($1.5 vs $8 per 1M).
MiniMax M3 costs 5.3× less per token ($1.5 vs $8 per 1M).
Task by task
| Task area | Grok 4.20 Beta | MiniMax M3 | Winner |
|---|---|---|---|
| Data & Analytics |
#91
/ 110
Excellent
|
#1
/ 110
Excellent
|
MiniMax M3 |
| Legal & HR |
#106
/ 110
Strong
|
#34
/ 110
Excellent
|
MiniMax M3 |
| Content & Brand |
#99
/ 124
Usable
|
#28
/ 124
Strong
|
MiniMax M3 |
| Landing Pages |
#60
/ 72
Usable
|
#2
/ 72
Strong
|
MiniMax M3 |
| Research & Competitive Analysis |
#86
/ 110
Needs editing
|
#33
/ 110
Strong
|
MiniMax M3 |
| Sales |
#46
/ 110
Usable
|
#2
/ 110
Strong
|
MiniMax M3 |
| Creative & Comedy | #92 / 110 | #55 / 110 | MiniMax M3 |
| Chef / Home Cooking |
#38
/ 126
Usable
|
#74
/ 126
Usable
|
Grok 4.20 Beta |
| Translation & Localization |
#90
/ 110
Strong
|
#55
/ 110
Excellent
|
MiniMax M3 |
| Training & Education |
#76
/ 110
Strong
|
#42
/ 110
Excellent
|
MiniMax M3 |
| Investor & Pitch |
#43
/ 66
Usable
|
#12
/ 66
Strong
|
MiniMax M3 |
| Executive Assistant |
#69
/ 112
Strong
|
#39
/ 112
Strong
|
MiniMax M3 |
| Coding |
#77
/ 115
Strong
|
#49
/ 115
Strong
|
MiniMax M3 |
| Product & Project Management |
#31
/ 110
Excellent
|
#3
/ 110
Excellent
|
MiniMax M3 |
| Presentations & Decks |
#56
/ 110
Strong
|
#29
/ 110
Excellent
|
MiniMax M3 |
| AI Strategy |
#49
/ 126
Strong
|
#27
/ 126
Strong
|
MiniMax M3 |
| Structured Output |
#43
/ 113
Excellent
|
#65
/ 113
Strong
|
Grok 4.20 Beta |
| RAG, Safety & Grounding |
#93
/ 113
Strong
|
#72
/ 113
Excellent
|
MiniMax M3 |
| Knowledge & Docs |
#86
/ 110
Usable
|
#67
/ 110
Usable
|
MiniMax M3 |
| Summarization & Meeting Notes |
#95
/ 110
Excellent
|
#79
/ 110
Excellent
|
MiniMax M3 |
| Frontend & Landing Pages |
#32
/ 109
Needs editing
|
#20
/ 109
Needs editing
|
MiniMax M3 |
| Customer Support |
#72
/ 113
Usable
|
#83
/ 113
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.
Same task, both models — judged
Both models answered the same test case; an independent judge graded each. Quotes are the judge's actual rationale.
Data & Analytics
Sessions vs visitors denominator (Cedar & Sage) (Metric Calculation)
“The model completely fails basic arithmetic, stating that 200 / 10,000 equals 2.86%. It mixes up the result of the visitor conversion rate (200/7000 = 2.86%) with the formula for the session conversion rate (200/10000 = 2.0%), resulting in a fundamentally incorrect and contradictory answer.”
“The model correctly calculates both conversion rates, explicitly states the denominators, shows the formulas, and provides a clear summary with a useful business insight.”
Frequently asked
Is Grok 4.20 Beta better than MiniMax M3?
Across 22 task areas we benchmarked, MiniMax M3 ranks higher in 19 and Grok 4.20 Beta in 3.
Which is cheaper, Grok 4.20 Beta or MiniMax M3?
MiniMax M3 costs 5.3× less per token ($1.5 vs $8 per 1M).
What is Grok 4.20 Beta better at?
Grok 4.20 Beta out-ranks MiniMax M3 at Chef / Home Cooking, Structured Output, Customer Support.
What is MiniMax M3 better at?
MiniMax M3 out-ranks Grok 4.20 Beta at Data & Analytics, Legal & HR, Content & Brand.
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