Noometry

Model comparison

GLM-4.6 vs GPT-5.3 Codex

GPT-5.3 Codex is the stronger model overall, scoring 45.8 to 41.4 on the Noometry Index. GLM-4.6 costs 4.8× less per token, which makes it the better buy when GPT-5.3 Codex's lead doesn't matter for your workload.

Last verified . 3 shared benchmarks.

GLM-4.6 Z.ai (Zhipu)

41.4

Rank #135 Confirmed

GPT-5.3 Codex OpenAI

45.8

Rank #69 Reported

Summary

  • They share 3 benchmarks with published results for both. GLM-4.6 scores higher in 0 categories and GPT-5.3 Codex in 2 categories; 2 gaps are clear of the uncertainty.
  • The widest gap is in agentic & tool use, where GPT-5.3 Codex leads 48.0 to 32.3.
  • The biggest single-benchmark swing is Terminal-Bench: 24.5% for GLM-4.6 and 78.4% for GPT-5.3 Codex.
  • GLM-4.6 is cheaper at $0.60 / $2.20 per million input/output tokens, against $1.75 / $14 for GPT-5.3 Codex.
  • GPT-5.3 Codex accepts more context: 400K tokens versus 205K.
  • GLM-4.6 has downloadable open weights; the other is API-only.

Side by side

GLM-4.6GPT-5.3 Codex
ProviderZ.ai (Zhipu)OpenAI
Noometry Index41.445.8
Released2025-09-302026-02-05
WeightsOpenProprietary
Context window205K400K
Max output131K128K
Input $ / M tokens$0.60$1.75
Output $ / M tokens$2.20$14
Results tracked298

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

Category by category

Coding GPT-5.3 Codex leads

GLM-4.6: 40.1 (#148), GPT-5.3 Codex: 48.6 (#56)

BenchmarkGLM-4.6GPT-5.3 Codex
LMArena WebDev13401409
ALE-Bench340.821,655
SWE-bench Verified—74.8%
SWE-bench Verified (bash only)55.4%—
SciCode38.4%—
WeirdML—79.3%
LMArena Coding1449—

Agentic & Tool Use GPT-5.3 Codex leads

GLM-4.6: 32.3 (#66), GPT-5.3 Codex: 48.0 (#9)

BenchmarkGLM-4.6GPT-5.3 Codex
Terminal-Bench24.5%78.4%
Berkeley Function Calling Leaderboard72.4%—
METR Time Horizons—74.5%
Vending-Bench 2—5,940

Reasoning Not comparable

GLM-4.6: 23.7 (#172), GPT-5.3 Codex: —

BenchmarkGLM-4.6GPT-5.3 Codex
Kagi LLM Benchmark47.4%—
CritPt1.1%—
LMArena Hard Prompts1440—
Epoch Capabilities Index—156.77

Math Not comparable

GLM-4.6: 39.1 (#111), GPT-5.3 Codex: —

BenchmarkGLM-4.6GPT-5.3 Codex
LMArena Math1432—
FrontierMath (Feb 2025 set)3.8%—
FrontierMath Tier 4 (v1)2.1%—

Knowledge Not comparable

GLM-4.6: 40.2 (#124), GPT-5.3 Codex: —

BenchmarkGLM-4.6GPT-5.3 Codex
Vectara Hallucination Rate9.5%—
LMArena Expert1431—

Multilingual Not comparable

GLM-4.6: 53.5 (#66), GPT-5.3 Codex: —

BenchmarkGLM-4.6GPT-5.3 Codex
LMArena Non-English1426—
LMArena Chinese1499—
LMArena French1459—
LMArena German1447—
LMArena Japanese1393—
LMArena Korean1400—
LMArena Russian1419—
LMArena Spanish1436—

Instruction Following Not comparable

GLM-4.6: 74.3 (#98), GPT-5.3 Codex: —

BenchmarkGLM-4.6GPT-5.3 Codex
LMArena Instruction Following1410—

Long Context Not comparable

GLM-4.6: 43.4 (#94), GPT-5.3 Codex: —

BenchmarkGLM-4.6GPT-5.3 Codex
LMArena Longer Query1422—

Writing & Preference Not comparable

GLM-4.6: 61.1 (#90), GPT-5.3 Codex: —

BenchmarkGLM-4.6GPT-5.3 Codex
LMArena Text1440—
LMArena Creative Writing1411—
EQ-Bench Creative Writing1411—
LMArena Multi-Turn1427—

Frequently asked questions

Is GLM-4.6 better than GPT-5.3 Codex?

GPT-5.3 Codex is the stronger model overall, scoring 45.8 to 41.4 on the Noometry Index. GLM-4.6 costs 4.8× less per token, which makes it the better buy when GPT-5.3 Codex's lead doesn't matter for your workload.

Which is cheaper, GLM-4.6 or GPT-5.3 Codex?

GLM-4.6 is cheaper. It lists at $0.60 per million input tokens and $2.20 per million output tokens; GPT-5.3 Codex lists at $1.75 and $14.

Is GLM-4.6 or GPT-5.3 Codex better for coding?

GPT-5.3 Codex scores higher on coding benchmarks: 48.6 versus 40.1 in the Noometry coding category.

Which has the bigger context window?

GPT-5.3 Codex does, with 400K tokens against 205K.

How many benchmarks do GLM-4.6 and GPT-5.3 Codex share?

3 benchmarks have published results for both models. GLM-4.6 has 29 scored results on Noometry and GPT-5.3 Codex has 8.

Related comparisons

Go deeper