Model comparison

GPT-5 Nano vs Grok 4.20 (Non-Reasoning)

Grok 4.20 (Non-Reasoning) is the stronger model overall, scoring 48.6 to 33.5 on the Noometry Index. GPT-5 Nano costs 11× 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. 35 shared benchmarks.

GPT-5 Nano OpenAI

33.5

Rank #241 Confirmed

Grok 4.20 (Non-Reasoning) xAI

48.6

Rank #54 Confirmed

Summary

They share 35 benchmarks with published results for both. GPT-5 Nano scores higher in 1 category and Grok 4.20 (Non-Reasoning) in 9 categories; 9 gaps are clear of the uncertainty.
The widest gap is in reasoning, where Grok 4.20 (Non-Reasoning) leads 52.3 to 16.3.
The biggest single-benchmark swing is ARC-AGI-1: 20.7% for GPT-5 Nano and 89.5% for Grok 4.20 (Non-Reasoning).
GPT-5 Nano is cheaper at $0.05 / $0.40 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 400K.

Side by side

GPT-5 Nano and Grok 4.20 (Non-Reasoning) specifications
	GPT-5 Nano	Grok 4.20 (Non-Reasoning)
Provider	OpenAI	xAI
Noometry Index	33.5	48.6
Released	2025-08-07	2026-02-17
Weights	Proprietary	Proprietary
Context window	400K	1M
Max output	128K	30K
Input $ / M tokens	$0.05	$1.25
Output $ / M tokens	$0.40	$2.50
Results tracked	49	46

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

Category by category

Coding Grok 4.20 (Non-Reasoning) leads

GPT-5 Nano: 33.6 (#254), Grok 4.20 (Non-Reasoning): 42.1 (#112)

Coding benchmarks
Benchmark	GPT-5 Nano	Grok 4.20 (Non-Reasoning)
WeirdML	38.1%	52.3%
LMArena Coding	1351	1459
ALE-Bench	718.67	1,150
SWE-bench Verified (bash only)	34.8%	—
LMArena WebDev	—	1375

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

GPT-5 Nano: 25.8 (#106), Grok 4.20 (Non-Reasoning): 34.4 (#46)

Agentic & Tool Use benchmarks
Benchmark	GPT-5 Nano	Grok 4.20 (Non-Reasoning)
Terminal-Bench	21.8%	57.3%
Berkeley Function Calling Leaderboard	51.5%	—
τ²-bench Banking	—	18%
LMArena Search	—	1189
Vending-Bench 2	—	4,663

Reasoning Grok 4.20 (Non-Reasoning) leads

GPT-5 Nano: 16.3 (#306), Grok 4.20 (Non-Reasoning): 52.3 (#32)

Reasoning benchmarks
Benchmark	GPT-5 Nano	Grok 4.20 (Non-Reasoning)
ARC-AGI-2	2.6%	65.1%
Kagi LLM Benchmark	62.2%	75%
ARC-AGI-1	20.7%	89.5%
Chess Puzzles	27%	24%
LMArena Hard Prompts	1328	1451
DTBench	62.7%	90.1%
LMCA	7.9%	38.7%
Epoch Capabilities Index	139.38	151.98
ForecastBench	59.1	61.4
NYT Connections (extended)	—	85.4%
Thematic Generalization	—	63.8%
Mystery Game Puzzles	9%	—

Math Grok 4.20 (Non-Reasoning) leads

GPT-5 Nano: 29.4 (#241), Grok 4.20 (Non-Reasoning): 48.2 (#65)

Math benchmarks
Benchmark	GPT-5 Nano	Grok 4.20 (Non-Reasoning)
FrontierMath (Tiers 1-3)	20%	44.9%
FrontierMath Tier 4	2.4%	17.1%
OTIS Mock AIME 2024-2025	81.1%	92.2%
ProofBench	12%	14%
LMArena Math	1317	1455
Omni-MATH	54.6%	—
MATH Level 5	95.2%	—
FrontierMath (Feb 2025 set)	8.3%	—
FrontierMath Tier 4 (v1)	2.1%	—

Knowledge Grok 4.20 (Non-Reasoning) leads

GPT-5 Nano: 35.9 (#178), Grok 4.20 (Non-Reasoning): 52.8 (#60)

Knowledge benchmarks
Benchmark	GPT-5 Nano	Grok 4.20 (Non-Reasoning)
GPQA Diamond	69.4%	89.3%
SimpleQA Verified	11.7%	30.2%
LMArena Expert	1321	1439
MMLU-Pro	77.8%	—
Vectara Hallucination Rate	10.5%	—
GPQA (HELM)	67.9%	—

Multimodal Grok 4.20 (Non-Reasoning) leads

GPT-5 Nano: 31.3 (#108), Grok 4.20 (Non-Reasoning): 33.3 (#98)

Multimodal benchmarks
Benchmark	GPT-5 Nano	Grok 4.20 (Non-Reasoning)
LMArena Vision	1159	1263
VPCT	37.2%	—
Blueprint-Bench 2	—	0%
LMArena Document	—	1416

Multilingual Grok 4.20 (Non-Reasoning) leads

GPT-5 Nano: 45.3 (#172), Grok 4.20 (Non-Reasoning): 54.5 (#40)

Multilingual benchmarks
Benchmark	GPT-5 Nano	Grok 4.20 (Non-Reasoning)
LMArena Non-English	1313	1441
LMArena Chinese	1356	1481
LMArena German	1327	1465
LMArena Japanese	1226	1449
LMArena Korean	1269	1417
LMArena Russian	1296	1458
LMArena Spanish	1360	1443
LMArena French	—	1476

Instruction Following Too close to call

GPT-5 Nano: 75.0 (#79), Grok 4.20 (Non-Reasoning): 74.8 (#83)

Instruction Following benchmarks
Benchmark	GPT-5 Nano	Grok 4.20 (Non-Reasoning)
LMArena Instruction Following	1306	1420
IFEval	93.2%	—

Long Context Grok 4.20 (Non-Reasoning) leads

GPT-5 Nano: 31.3 (#281), Grok 4.20 (Non-Reasoning): 45.5 (#34)

Long Context benchmarks
Benchmark	GPT-5 Nano	Grok 4.20 (Non-Reasoning)
LMArena Longer Query	1312	1437
Fiction.LiveBench	44.4%	—
CL-bench	—	22.2%
CL-bench Life	—	11.9%

Writing & Preference Grok 4.20 (Non-Reasoning) leads

GPT-5 Nano: 39.1 (#249), Grok 4.20 (Non-Reasoning): 65.7 (#44)

Writing & Preference benchmarks
Benchmark	GPT-5 Nano	Grok 4.20 (Non-Reasoning)
LMArena Text	1320	1451
LMArena Creative Writing	1249	1438
EQ-Bench Creative Writing	705	1574
LMArena Multi-Turn	1311	1456
WildBench	80.6%	—

Frequently asked questions

Is GPT-5 Nano better than Grok 4.20 (Non-Reasoning)?

Which is cheaper, GPT-5 Nano or Grok 4.20 (Non-Reasoning)?

GPT-5 Nano is cheaper. It lists at $0.05 per million input tokens and $0.40 per million output tokens; Grok 4.20 (Non-Reasoning) lists at $1.25 and $2.50.

Is GPT-5 Nano or Grok 4.20 (Non-Reasoning) better for coding?

Grok 4.20 (Non-Reasoning) scores higher on coding benchmarks: 42.1 versus 33.6 in the Noometry coding category.

Which has the bigger context window?

Grok 4.20 (Non-Reasoning) does, with 1M tokens against 400K.

How many benchmarks do GPT-5 Nano and Grok 4.20 (Non-Reasoning) share?

35 benchmarks have published results for both models. GPT-5 Nano has 49 scored results on Noometry and Grok 4.20 (Non-Reasoning) has 46.