Model comparison

Grok 4.20 (Non-Reasoning) vs MiniMax-M2.7

Grok 4.20 (Non-Reasoning) is the stronger model overall, scoring 48.6 to 37.7 on the Noometry Index. MiniMax-M2.7 costs 3.0× less per token, which makes it the better buy when Grok 4.20 (Non-Reasoning)'s lead doesn't matter for your workload.

Last verified October 10, 2026. 25 shared benchmarks.

Grok 4.20 (Non-Reasoning) xAI

48.6

Rank #54 Confirmed

MiniMax-M2.7

MiniMax

37.7

Rank #196 Confirmed

Summary

They share 25 benchmarks with published results for both. Grok 4.20 (Non-Reasoning) scores higher in 9 categories and MiniMax-M2.7 in 0 categories; 7 gaps are clear of the uncertainty.
The widest gap is in reasoning, where Grok 4.20 (Non-Reasoning) leads 52.3 to 19.7.
The biggest single-benchmark swing is NYT Connections (extended): 85.4% for Grok 4.20 (Non-Reasoning) and 24.7% for MiniMax-M2.7.
MiniMax-M2.7 is cheaper at $0.30 / $1.20 per million input/output tokens, against $1.25 / $2.50 for Grok 4.20 (Non-Reasoning).
Grok 4.20 (Non-Reasoning) accepts more context: 1M tokens versus 205K.
MiniMax-M2.7 has downloadable open weights; the other is API-only.

Side by side

Grok 4.20 (Non-Reasoning) and MiniMax-M2.7 specifications
	Grok 4.20 (Non-Reasoning)	MiniMax-M2.7
Provider	xAI	MiniMax
Noometry Index	48.6	37.7
Released	2026-02-17	2026-03-18
Weights	Proprietary	Open
Context window	1M	205K
Max output	30K	131K
Input $ / M tokens	$1.25	$0.30
Output $ / M tokens	$2.50	$1.20
Results tracked	46	30

Sponsored placements are available on pages like this one. Advertise on Noometry

Category by category

Coding Too close to call

Grok 4.20 (Non-Reasoning): 42.1 (#112), MiniMax-M2.7: 41.8 (#120)

Coding benchmarks
Benchmark	Grok 4.20 (Non-Reasoning)	MiniMax-M2.7
LMArena WebDev	1375	1398
WeirdML	52.3%	37%
LMArena Coding	1459	1454
ALE-Bench	1,150	599.25
SciCode	—	47%

Agentic & Tool Use Grok 4.20 (Non-Reasoning) leads

Grok 4.20 (Non-Reasoning): 34.4 (#46), MiniMax-M2.7: 25.1 (#111)

Agentic & Tool Use benchmarks
Benchmark	Grok 4.20 (Non-Reasoning)	MiniMax-M2.7
Terminal-Bench	57.3%	45.1%
τ²-bench Banking	18%	—
ExploitBench	—	13.3%
GBAEval	—	0%
LMArena Search	1189	—
Vending-Bench 2	4,663	—

Reasoning Grok 4.20 (Non-Reasoning) leads

Grok 4.20 (Non-Reasoning): 52.3 (#32), MiniMax-M2.7: 19.7 (#253)

Reasoning benchmarks
Benchmark	Grok 4.20 (Non-Reasoning)	MiniMax-M2.7
NYT Connections (extended)	85.4%	24.7%
Thematic Generalization	63.8%	39.3%
LMArena Hard Prompts	1451	1422
Epoch Capabilities Index	151.98	145.85
ARC-AGI-2	65.1%	—
Kagi LLM Benchmark	75%	—
ARC-AGI-1	89.5%	—
CritPt	—	0.6%
Chess Puzzles	24%	—
DTBench	90.1%	—
LMCA	38.7%	—
ForecastBench	61.4	—

Math Grok 4.20 (Non-Reasoning) leads

Grok 4.20 (Non-Reasoning): 48.2 (#65), MiniMax-M2.7: 25.9 (#263)

Math benchmarks
Benchmark	Grok 4.20 (Non-Reasoning)	MiniMax-M2.7
ProofBench	14%	3%
LMArena Math	1455	1420
FrontierMath (Tiers 1-3)	44.9%	—
FrontierMath Tier 4	17.1%	—
OTIS Mock AIME 2024-2025	92.2%	—

Knowledge Grok 4.20 (Non-Reasoning) leads

Grok 4.20 (Non-Reasoning): 52.8 (#60), MiniMax-M2.7: 37.7 (#152)

Knowledge benchmarks
Benchmark	Grok 4.20 (Non-Reasoning)	MiniMax-M2.7
LMArena Expert	1439	1444
GPQA Diamond	89.3%	—
SimpleQA Verified	30.2%	—
Vectara Hallucination Rate	—	12.9%

Multimodal Not comparable

Grok 4.20 (Non-Reasoning): 33.3 (#98), MiniMax-M2.7: —

Multimodal benchmarks
Benchmark	Grok 4.20 (Non-Reasoning)	MiniMax-M2.7
LMArena Vision	1263	—
Blueprint-Bench 2	0%	—
LMArena Document	1416	—

Multilingual Grok 4.20 (Non-Reasoning) leads

Grok 4.20 (Non-Reasoning): 54.5 (#40), MiniMax-M2.7: 50.3 (#123)

Multilingual benchmarks
Benchmark	Grok 4.20 (Non-Reasoning)	MiniMax-M2.7
LMArena Non-English	1441	1382
LMArena Chinese	1481	1441
LMArena French	1476	1421
LMArena German	1465	1398
LMArena Japanese	1449	1262
LMArena Korean	1417	1313
LMArena Russian	1458	1383
LMArena Spanish	1443	1403

Instruction Following Too close to call

Grok 4.20 (Non-Reasoning): 74.8 (#83), MiniMax-M2.7: 74.1 (#103)

Instruction Following benchmarks
Benchmark	Grok 4.20 (Non-Reasoning)	MiniMax-M2.7
LMArena Instruction Following	1420	1405

Long Context Grok 4.20 (Non-Reasoning) leads

Grok 4.20 (Non-Reasoning): 45.5 (#34), MiniMax-M2.7: 43.3 (#99)

Long Context benchmarks
Benchmark	Grok 4.20 (Non-Reasoning)	MiniMax-M2.7
LMArena Longer Query	1437	1419
CL-bench	22.2%	—
CL-bench Life	11.9%	—

Writing & Preference Grok 4.20 (Non-Reasoning) leads

Grok 4.20 (Non-Reasoning): 65.7 (#44), MiniMax-M2.7: 58.9 (#112)

Writing & Preference benchmarks
Benchmark	Grok 4.20 (Non-Reasoning)	MiniMax-M2.7
LMArena Text	1451	1405
LMArena Creative Writing	1438	1354
LMArena Multi-Turn	1456	1412
EQ-Bench Creative Writing	1574	—

Frequently asked questions

Is Grok 4.20 (Non-Reasoning) better than MiniMax-M2.7?

Which is cheaper, Grok 4.20 (Non-Reasoning) or MiniMax-M2.7?

MiniMax-M2.7 is cheaper. It lists at $0.30 per million input tokens and $1.20 per million output tokens; Grok 4.20 (Non-Reasoning) lists at $1.25 and $2.50.

Is Grok 4.20 (Non-Reasoning) or MiniMax-M2.7 better for coding?

They score almost the same on coding (42.1 vs 41.8); test both on your own repository before choosing.

Which has the bigger context window?

Grok 4.20 (Non-Reasoning) does, with 1M tokens against 205K.

How many benchmarks do Grok 4.20 (Non-Reasoning) and MiniMax-M2.7 share?

25 benchmarks have published results for both models. Grok 4.20 (Non-Reasoning) has 46 scored results on Noometry and MiniMax-M2.7 has 30.