Noometry

Model comparison

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

Grok 4.20 (Non-Reasoning) is the stronger model overall, scoring 48.5 to 41.7 on the Noometry Index. GPT-5 Mini costs 2.3× 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 . 37 shared benchmarks.

GPT-5 Mini OpenAI

41.7

Rank #115 Confirmed

Grok 4.20 (Non-Reasoning) xAI

48.5

Rank #48 Confirmed

Summary

  • They share 37 benchmarks with published results for both. GPT-5 Mini scores higher in 2 categories and Grok 4.20 (Non-Reasoning) in 8 categories; 10 gaps are clear of the uncertainty.
  • The widest gap is in reasoning, where Grok 4.20 (Non-Reasoning) leads 52.2 to 24.1.
  • The biggest single-benchmark swing is ARC-AGI-2: 4.4% for GPT-5 Mini and 65.1% for Grok 4.20 (Non-Reasoning).
  • GPT-5 Mini is cheaper at $0.25 / $2 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 MiniGrok 4.20 (Non-Reasoning)
ProviderOpenAIxAI
Noometry Index41.748.5
Released2025-08-072026-02-17
WeightsProprietaryProprietary
Context window400K1M
Max output128K30K
Input $ / M tokens$0.25$1.25
Output $ / M tokens$2$2.50
Results tracked6046

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

Category by category

Coding Grok 4.20 (Non-Reasoning) leads

GPT-5 Mini: 40.0 (#134), Grok 4.20 (Non-Reasoning): 41.8 (#104)

BenchmarkGPT-5 MiniGrok 4.20 (Non-Reasoning)
WeirdML52.7%52.3%
LMArena Coding14061458
ALE-Bench799.771,150
SWE-bench Verified64.7%—
SWE-bench Verified (bash only)59.8%—
LMArena WebDev—1374
SWE-bench Multilingual39.7%—
SciCode39.2%—
AlgoTune1.38—

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

GPT-5 Mini: 31.0 (#64), Grok 4.20 (Non-Reasoning): 34.3 (#43)

BenchmarkGPT-5 MiniGrok 4.20 (Non-Reasoning)
Terminal-Bench34.8%57.3%
Vending-Bench 2-31.184,663
Berkeley Function Calling Leaderboard55.5%—
τ²-bench Banking—18%
LMArena Search—1189

Reasoning Grok 4.20 (Non-Reasoning) leads

GPT-5 Mini: 24.1 (#153), Grok 4.20 (Non-Reasoning): 52.2 (#30)

BenchmarkGPT-5 MiniGrok 4.20 (Non-Reasoning)
ARC-AGI-24.4%65.1%
Kagi LLM Benchmark70.3%75%
ARC-AGI-154.3%89.5%
Chess Puzzles30%24%
LMArena Hard Prompts13801451
DTBench80.5%90.1%
LMCA34.2%38.7%
Epoch Capabilities Index145.51152.01
ForecastBench6161.4
NYT Connections (extended)—85.4%
CritPt0%—
EnigmaEval8.2%—
Thematic Generalization—63.8%
Mystery Game Puzzles10%—

Math Grok 4.20 (Non-Reasoning) leads

GPT-5 Mini: 46.3 (#59), Grok 4.20 (Non-Reasoning): 48.0 (#54)

BenchmarkGPT-5 MiniGrok 4.20 (Non-Reasoning)
FrontierMath (Tiers 1-3)46.7%44.9%
FrontierMath Tier 412.2%17.1%
OTIS Mock AIME 2024-202586.7%92.2%
ProofBench9%14%
LMArena Math13771455
Omni-MATH72.2%—
MATH Level 597.8%—
FrontierMath (Feb 2025 set)27.2%—
FrontierMath Tier 4 (v1)6.3%—

Knowledge Grok 4.20 (Non-Reasoning) leads

GPT-5 Mini: 45.5 (#80), Grok 4.20 (Non-Reasoning): 52.8 (#54)

BenchmarkGPT-5 MiniGrok 4.20 (Non-Reasoning)
GPQA Diamond75%89.3%
SimpleQA Verified21.6%30.2%
LMArena Expert13771440
Humanity's Last Exam19.4%—
MMLU-Pro83.5%—
Confabulations13.3%—
Vectara Hallucination Rate12.9%—
GPQA (HELM)75.6%—

Multimodal GPT-5 Mini leads

GPT-5 Mini: 35.0 (#80), Grok 4.20 (Non-Reasoning): 33.2 (#88)

BenchmarkGPT-5 MiniGrok 4.20 (Non-Reasoning)
LMArena Vision12021263
VPCT40.2%—
Blueprint-Bench 2—0%
LMArena Document—1416

Multilingual Grok 4.20 (Non-Reasoning) leads

GPT-5 Mini: 49.1 (#131), Grok 4.20 (Non-Reasoning): 54.7 (#38)

BenchmarkGPT-5 MiniGrok 4.20 (Non-Reasoning)
LMArena Non-English13631440
LMArena Chinese13871481
LMArena French13851477
LMArena German13661463
LMArena Japanese13421450
LMArena Korean13061417
LMArena Russian13621457
LMArena Spanish13551442

Instruction Following GPT-5 Mini leads

GPT-5 Mini: 76.4 (#43), Grok 4.20 (Non-Reasoning): 75.0 (#79)

BenchmarkGPT-5 MiniGrok 4.20 (Non-Reasoning)
LMArena Instruction Following13571420
IFEval92.7%—

Long Context Grok 4.20 (Non-Reasoning) leads

GPT-5 Mini: 41.9 (#127), Grok 4.20 (Non-Reasoning): 45.4 (#33)

BenchmarkGPT-5 MiniGrok 4.20 (Non-Reasoning)
LMArena Longer Query13551436
Fiction.LiveBench69.4%—
CL-bench—22.2%
CL-bench Life—11.9%

Writing & Preference Grok 4.20 (Non-Reasoning) leads

GPT-5 Mini: 55.4 (#143), Grok 4.20 (Non-Reasoning): 65.9 (#42)

BenchmarkGPT-5 MiniGrok 4.20 (Non-Reasoning)
LMArena Text13731451
LMArena Creative Writing13251439
EQ-Bench Creative Writing13131574
LMArena Multi-Turn13631456
Short-Story Creative Writing83.1%—
WildBench85.5%—

Frequently asked questions

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

Grok 4.20 (Non-Reasoning) is the stronger model overall, scoring 48.5 to 41.7 on the Noometry Index. GPT-5 Mini costs 2.3× less per token, which makes it the better buy when Grok 4.20 (Non-Reasoning)'s lead doesn't matter for your workload.

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

GPT-5 Mini is cheaper. It lists at $0.25 per million input tokens and $2 per million output tokens; Grok 4.20 (Non-Reasoning) lists at $1.25 and $2.50.

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

Grok 4.20 (Non-Reasoning) scores higher on coding benchmarks: 41.8 versus 40.0 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 Mini and Grok 4.20 (Non-Reasoning) share?

37 benchmarks have published results for both models. GPT-5 Mini has 60 scored results on Noometry and Grok 4.20 (Non-Reasoning) has 46.

Related comparisons

Go deeper