Noometry

Model comparison

GLM-5.3-Flash vs GPT-5.4 mini

GLM-5.3-Flash is the stronger model overall, scoring 51.8 to 45.0 on the Noometry Index.

Last verified . 33 shared benchmarks.

GLM-5.3-Flash Z.ai (Zhipu)

51.8

Rank #41 Confirmed

GPT-5.4 mini OpenAI

45.0

Rank #76 Confirmed

Summary

  • They share 33 benchmarks with published results for both. GLM-5.3-Flash scores higher in 10 categories and GPT-5.4 mini in 0 categories; 10 gaps are clear of the uncertainty.
  • The widest gap is in reasoning, where GLM-5.3-Flash leads 48.0 to 30.4.
  • The biggest single-benchmark swing is ARC-AGI-2: 65.8% for GLM-5.3-Flash and 18.9% for GPT-5.4 mini.
  • GLM-5.3-Flash is cheaper at $0.15 / $0.50 per million input/output tokens, against $0.75 / $4.50 for GPT-5.4 mini.
  • GLM-5.3-Flash accepts more context: 1M tokens versus 400K.
  • GLM-5.3-Flash has downloadable open weights; the other is API-only.

Side by side

GLM-5.3-FlashGPT-5.4 mini
ProviderZ.ai (Zhipu)OpenAI
Noometry Index51.845.0
Released2026-08-202026-03-17
WeightsOpenProprietary
Context window1M400K
Max output131K128K
Input $ / M tokens$0.15$0.75
Output $ / M tokens$0.50$4.50
Results tracked4046

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

Category by category

Coding GLM-5.3-Flash leads

GLM-5.3-Flash: 53.1 (#31), GPT-5.4 mini: 45.2 (#72)

BenchmarkGLM-5.3-FlashGPT-5.4 mini
FrontierCode31.8%27%
LMArena WebDev16091397
SciCode51.6%49.9%
LMArena Coding15081438
ALE-Bench303.551,189
DeepSWE63.4%—
CursorBench36.8%—
FrontierSWE18.1%—
WeirdML—60.3%

Agentic & Tool Use GLM-5.3-Flash leads

GLM-5.3-Flash: 34.2 (#47), GPT-5.4 mini: 29.9 (#81)

BenchmarkGLM-5.3-FlashGPT-5.4 mini
APEX-Agents52.8%—
DeepResearch Bench—36.3%
GDP.pdf14%—

Reasoning GLM-5.3-Flash leads

GLM-5.3-Flash: 48.0 (#42), GPT-5.4 mini: 30.4 (#85)

BenchmarkGLM-5.3-FlashGPT-5.4 mini
ARC-AGI-265.8%18.9%
ARC-AGI-191%63.7%
CritPt15.4%10%
Chess Puzzles14%24%
LMArena Hard Prompts14911424
Mystery Game Puzzles8%11%
Epoch Capabilities Index151.88148.84
Kagi LLM Benchmark—37.9%
NYT Connections (extended)—61.8%
Thematic Generalization—61.7%
DTBench—80%
LMCA—40.8%
Surface Evolver Bench52.5%—
Bench to the Future 30.15—
ForecastBench—57

Math GLM-5.3-Flash leads

GLM-5.3-Flash: 53.3 (#47), GPT-5.4 mini: 45.5 (#75)

BenchmarkGLM-5.3-FlashGPT-5.4 mini
FrontierMath (Tiers 1-3)55.8%51.2%
FrontierMath Tier 417.1%9.8%
OTIS Mock AIME 2024-202593.9%88.9%
ProofBench21%21%
LMArena Math15001419
FrontierMath (Feb 2025 set)—28.3%
FrontierMath Tier 4 (v1)—2.1%

Knowledge GLM-5.3-Flash leads

GLM-5.3-Flash: 58.4 (#36), GPT-5.4 mini: 51.5 (#67)

BenchmarkGLM-5.3-FlashGPT-5.4 mini
GPQA Diamond90.2%86.9%
LMArena Expert15131435
SimpleQA Verified—29.4%
Vectara Hallucination Rate—5.5%

Multimodal GLM-5.3-Flash leads

GLM-5.3-Flash: 42.8 (#27), GPT-5.4 mini: 39.7 (#56)

BenchmarkGLM-5.3-FlashGPT-5.4 mini
LMArena Vision12961245

Multilingual GLM-5.3-Flash leads

GLM-5.3-Flash: 56.0 (#25), GPT-5.4 mini: 51.9 (#96)

BenchmarkGLM-5.3-FlashGPT-5.4 mini
LMArena Non-English14621405
LMArena Chinese15271446
LMArena French14961440
LMArena German14701409
LMArena Japanese14291374
LMArena Korean14461368
LMArena Russian14691417
LMArena Spanish14711405

Instruction Following GLM-5.3-Flash leads

GLM-5.3-Flash: 77.5 (#20), GPT-5.4 mini: 74.1 (#102)

BenchmarkGLM-5.3-FlashGPT-5.4 mini
LMArena Instruction Following14781405

Long Context GLM-5.3-Flash leads

GLM-5.3-Flash: 45.4 (#39), GPT-5.4 mini: 43.0 (#112)

BenchmarkGLM-5.3-FlashGPT-5.4 mini
LMArena Longer Query14821407

Writing & Preference GLM-5.3-Flash leads

GLM-5.3-Flash: 65.3 (#50), GPT-5.4 mini: 64.0 (#58)

BenchmarkGLM-5.3-FlashGPT-5.4 mini
LMArena Text14711412
LMArena Creative Writing14421370
LMArena Multi-Turn14671429
EQ-Bench Creative Writing—1665

Frequently asked questions

Is GLM-5.3-Flash better than GPT-5.4 mini?

GLM-5.3-Flash is the stronger model overall, scoring 51.8 to 45.0 on the Noometry Index.

Which is cheaper, GLM-5.3-Flash or GPT-5.4 mini?

GLM-5.3-Flash is cheaper. It lists at $0.15 per million input tokens and $0.50 per million output tokens; GPT-5.4 mini lists at $0.75 and $4.50.

Is GLM-5.3-Flash or GPT-5.4 mini better for coding?

GLM-5.3-Flash scores higher on coding benchmarks: 53.1 versus 45.2 in the Noometry coding category.

Which has the bigger context window?

GLM-5.3-Flash does, with 1M tokens against 400K.

How many benchmarks do GLM-5.3-Flash and GPT-5.4 mini share?

33 benchmarks have published results for both models. GLM-5.3-Flash has 40 scored results on Noometry and GPT-5.4 mini has 46.

Related comparisons

Go deeper