Noometry

Model comparison

GLM-4.5V vs GPT-4

GLM-4.5V is the stronger model overall, scoring 39.8 to 29.1 on the Noometry Index.

Last verified . 13 shared benchmarks.

GLM-4.5V Z.ai (Zhipu)

39.8

Rank #158 Confirmed

GPT-4 OpenAI

29.1

Rank #316 Confirmed

Summary

  • They share 13 benchmarks with published results for both. GLM-4.5V scores higher in 8 categories and GPT-4 in 0 categories; 8 gaps are clear of the uncertainty.
  • The widest gap is in math, where GLM-4.5V leads 37.4 to 10.8.
  • GLM-4.5V is cheaper at $0.60 / $1.80 per million input/output tokens, against $30 / $60 for GPT-4.
  • GLM-4.5V accepts more context: 64K tokens versus 8K.
  • GLM-4.5V has downloadable open weights; the other is API-only.

Side by side

GLM-4.5VGPT-4
ProviderZ.ai (Zhipu)OpenAI
Noometry Index39.829.1
Released2025-08-112023-03-14
WeightsOpenProprietary
Context window64K8K
Max output16K8K
Input $ / M tokens$0.60$30
Output $ / M tokens$1.80$60
Results tracked1538

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

Category by category

Coding GLM-4.5V leads

GLM-4.5V: 39.5 (#155), GPT-4: 31.6 (#283)

BenchmarkGLM-4.5VGPT-4
LMArena Coding13471254
WeirdML—12.4%
BigCodeBench Instruct—46%
BigCodeBench Complete—57.2%
HumanEval+—79.3%

Agentic & Tool Use Not comparable

GLM-4.5V: —, GPT-4: —

BenchmarkGLM-4.5VGPT-4
METR Time Horizons—36.1%

Reasoning GLM-4.5V leads

GLM-4.5V: 27.4 (#119), GPT-4: 17.8 (#289)

BenchmarkGLM-4.5VGPT-4
LMArena Hard Prompts13341241
Kagi LLM Benchmark59.8%—
Chess Puzzles—4%
Mystery Game Puzzles—12%
DTBench—62.7%
LMCA—17.1%
BIG-Bench Hard—75.1%
Epoch Capabilities Index—125.89
ForecastBench—57.8
HellaSwag—95.3%
WinoGrande—87.5%

Math GLM-4.5V leads

GLM-4.5V: 37.4 (#159), GPT-4: 10.8 (#309)

BenchmarkGLM-4.5VGPT-4
LMArena Math13541269
OTIS Mock AIME 2024-2025—1.1%
MATH Level 5—23%
GSM8K—92%

Knowledge GLM-4.5V leads

GLM-4.5V: 37.5 (#156), GPT-4: 18.4 (#282)

BenchmarkGLM-4.5VGPT-4
LMArena Expert13531211
GPQA Diamond—35.7%
MMLU—86.4%
TriviaQA—84.8%

Multimodal Not comparable

GLM-4.5V: 34.3 (#92), GPT-4: —

BenchmarkGLM-4.5VGPT-4
LMArena Vision1154—

Multilingual GLM-4.5V leads

GLM-4.5V: 44.6 (#177), GPT-4: 40.6 (#215)

BenchmarkGLM-4.5VGPT-4
LMArena Non-English13031246
LMArena Chinese13371242
LMArena Russian12981251
LMArena Spanish13361261
LMArena French—1283
LMArena German—1251
LMArena Japanese—1209
LMArena Korean—1184

Instruction Following GLM-4.5V leads

GLM-4.5V: 69.2 (#175), GPT-4: 65.3 (#222)

BenchmarkGLM-4.5VGPT-4
LMArena Instruction Following13111241

Long Context GLM-4.5V leads

GLM-4.5V: 39.6 (#171), GPT-4: 37.7 (#212)

BenchmarkGLM-4.5VGPT-4
LMArena Longer Query13041244

Writing & Preference GLM-4.5V leads

GLM-4.5V: 52.5 (#170), GPT-4: 34.9 (#268)

BenchmarkGLM-4.5VGPT-4
LMArena Text13331263
LMArena Creative Writing12951244
LMArena Multi-Turn13321257
EQ-Bench Creative Writing—752

Frequently asked questions

Is GLM-4.5V better than GPT-4?

GLM-4.5V is the stronger model overall, scoring 39.8 to 29.1 on the Noometry Index.

Which is cheaper, GLM-4.5V or GPT-4?

GLM-4.5V is cheaper. It lists at $0.60 per million input tokens and $1.80 per million output tokens; GPT-4 lists at $30 and $60.

Is GLM-4.5V or GPT-4 better for coding?

GLM-4.5V scores higher on coding benchmarks: 39.5 versus 31.6 in the Noometry coding category.

Which has the bigger context window?

GLM-4.5V does, with 64K tokens against 8K.

How many benchmarks do GLM-4.5V and GPT-4 share?

13 benchmarks have published results for both models. GLM-4.5V has 15 scored results on Noometry and GPT-4 has 38.

Related comparisons

Go deeper