Noometry

Model comparison

GLM-5.3 vs Grok-3 mini

GLM-5.3 is the stronger model overall, scoring 54.8 to 41.2 on the Noometry Index.

Last verified . 21 shared benchmarks.

GLM-5.3 Z.ai (Zhipu)

54.8

Rank #26 Confirmed

Grok-3 mini xAI

41.2

Rank #141 Confirmed

Summary

  • They share 21 benchmarks with published results for both. GLM-5.3 scores higher in 7 categories and Grok-3 mini in 1 category; 8 gaps are clear of the uncertainty.
  • The widest gap is in reasoning, where GLM-5.3 leads 46.1 to 13.6.
  • The biggest single-benchmark swing is WeirdML: 75.4% for GLM-5.3 and 42.6% for Grok-3 mini.
  • GLM-5.3 has downloadable open weights; the other is API-only.

Side by side

GLM-5.3Grok-3 mini
ProviderZ.ai (Zhipu)xAI
Noometry Index54.841.2
Released2026-08-142025-04-09
WeightsOpenProprietary
Context window1M—
Max output131K—
Input $ / M tokens$1.40—
Output $ / M tokens$4.40—
Results tracked4235

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

Category by category

Coding GLM-5.3 leads

GLM-5.3: 59.5 (#14), Grok-3 mini: 40.8 (#131)

BenchmarkGLM-5.3Grok-3 mini
WeirdML75.4%42.6%
LMArena Coding14961379
DeepSWE69%—
FrontierCode40.1%—
Aider Polyglot—49.3%
CursorBench42.6%—
LMArena WebDev1622—
FrontierSWE30.2%—
SciCode59%—
ALE-Bench1,317—

Agentic & Tool Use Not comparable

GLM-5.3: 36.4 (#38), Grok-3 mini: —

BenchmarkGLM-5.3Grok-3 mini
APEX-Agents56.6%—
Vending-Bench 28,164—

Reasoning GLM-5.3 leads

GLM-5.3: 46.1 (#46), Grok-3 mini: 13.6 (#334)

BenchmarkGLM-5.3Grok-3 mini
LMArena Hard Prompts14891375
Epoch Capabilities Index155.61140.35
ARC-AGI-2—0.4%
Kagi LLM Benchmark—61.3%
NYT Connections (extended)74.2%—
ARC-AGI-1—16.5%
CritPt19.1%—
Chess Puzzles21%—
Mystery Game Puzzles33%—
DTBench87.7%—
LMCA55.5%—
Bench to the Future 30.15—

Math GLM-5.3 leads

GLM-5.3: 62.3 (#33), Grok-3 mini: 42.1 (#85)

BenchmarkGLM-5.3Grok-3 mini
OTIS Mock AIME 2024-202591.1%77.8%
LMArena Math14891386
FrontierMath (Tiers 1-3)68.8%—
FrontierMath Tier 429.3%—
ProofBench49%—
Omni-MATH—31.8%
MATH Level 5—90.9%
FrontierMath (Feb 2025 set)—5.9%

Knowledge GLM-5.3 leads

GLM-5.3: 58.3 (#37), Grok-3 mini: 46.4 (#81)

BenchmarkGLM-5.3Grok-3 mini
GPQA Diamond90.9%76.3%
LMArena Expert15161395
SimpleQA Verified41%—
MMLU-Pro—79.9%
Confabulations—10.8%
GPQA (HELM)—67.5%

Multilingual GLM-5.3 leads

GLM-5.3: 55.7 (#28), Grok-3 mini: 48.1 (#145)

BenchmarkGLM-5.3Grok-3 mini
LMArena Non-English14571352
LMArena Chinese15281387
LMArena French14991357
LMArena German14991349
LMArena Japanese14531342
LMArena Korean14721335
LMArena Russian14631353
LMArena Spanish14601381

Instruction Following Grok-3 mini leads

GLM-5.3: 77.5 (#23), Grok-3 mini: 78.5 (#9)

BenchmarkGLM-5.3Grok-3 mini
LMArena Instruction Following14771357
IFEval—95.1%

Long Context GLM-5.3 leads

GLM-5.3: 45.4 (#41), Grok-3 mini: 41.0 (#147)

BenchmarkGLM-5.3Grok-3 mini
LMArena Longer Query14821372
Fiction.LiveBench—66.7%

Writing & Preference GLM-5.3 leads

GLM-5.3: 75.7 (#6), Grok-3 mini: 52.5 (#169)

BenchmarkGLM-5.3Grok-3 mini
LMArena Text14711370
LMArena Creative Writing14571342
LMArena Multi-Turn14721355
Short-Story Creative Writing—73.5%
EQ-Bench Creative Writing2075—
WildBench—65.1%

Frequently asked questions

Is GLM-5.3 better than Grok-3 mini?

GLM-5.3 is the stronger model overall, scoring 54.8 to 41.2 on the Noometry Index.

Is GLM-5.3 or Grok-3 mini better for coding?

GLM-5.3 scores higher on coding benchmarks: 59.5 versus 40.8 in the Noometry coding category.

How many benchmarks do GLM-5.3 and Grok-3 mini share?

21 benchmarks have published results for both models. GLM-5.3 has 42 scored results on Noometry and Grok-3 mini has 35.

Related comparisons

Go deeper