Noometry

Model comparison

GLM-5.3 vs Hy3

GLM-5.3 is the stronger model overall, scoring 54.8 to 44.2 on the Noometry Index. Hy3 costs 15× less per token, which makes it the better buy when GLM-5.3's lead doesn't matter for your workload.

Last verified . 19 shared benchmarks.

GLM-5.3 Z.ai (Zhipu)

54.8

Rank #26 Confirmed

Hy3 Tencent

44.2

Rank #79 Confirmed

Summary

  • They share 19 benchmarks with published results for both. GLM-5.3 scores higher in 8 categories and Hy3 in 0 categories; 8 gaps are clear of the uncertainty.
  • The widest gap is in math, where GLM-5.3 leads 62.3 to 40.1.
  • The biggest single-benchmark swing is NYT Connections (extended): 74.2% for GLM-5.3 and 41.2% for Hy3.
  • Hy3 is cheaper at $0.0825 / $0.33 per million input/output tokens, against $1.40 / $4.40 for GLM-5.3.
  • GLM-5.3 accepts more context: 1M tokens versus 262K.

Side by side

GLM-5.3Hy3
ProviderZ.ai (Zhipu)Tencent
Noometry Index54.844.2
Released2026-08-142026-07-06
WeightsOpenOpen
Context window1M262K
Max output131K128K
Input $ / M tokens$1.40$0.0825
Output $ / M tokens$4.40$0.33
Results tracked4219

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), Hy3: 46.8 (#63)

BenchmarkGLM-5.3Hy3
LMArena WebDev16221508
LMArena Coding14961464
DeepSWE69%—
FrontierCode40.1%—
CursorBench42.6%—
FrontierSWE30.2%—
SciCode59%—
WeirdML75.4%—
ALE-Bench1,317—

Agentic & Tool Use Not comparable

GLM-5.3: 36.4 (#38), Hy3: —

BenchmarkGLM-5.3Hy3
APEX-Agents56.6%—
Vending-Bench 28,164—

Reasoning GLM-5.3 leads

GLM-5.3: 46.1 (#46), Hy3: 26.1 (#136)

BenchmarkGLM-5.3Hy3
NYT Connections (extended)74.2%41.2%
LMArena Hard Prompts14891447
CritPt19.1%—
Chess Puzzles21%—
Mystery Game Puzzles33%—
DTBench87.7%—
LMCA55.5%—
Bench to the Future 30.15—
Epoch Capabilities Index155.61—

Math GLM-5.3 leads

GLM-5.3: 62.3 (#33), Hy3: 40.1 (#93)

BenchmarkGLM-5.3Hy3
LMArena Math14891475
FrontierMath (Tiers 1-3)68.8%—
FrontierMath Tier 429.3%—
OTIS Mock AIME 2024-202591.1%—
ProofBench49%—

Knowledge GLM-5.3 leads

GLM-5.3: 58.3 (#37), Hy3: 40.8 (#114)

BenchmarkGLM-5.3Hy3
LMArena Expert15161460
GPQA Diamond90.9%—
SimpleQA Verified41%—

Multilingual GLM-5.3 leads

GLM-5.3: 55.7 (#28), Hy3: 53.5 (#65)

BenchmarkGLM-5.3Hy3
LMArena Non-English14571426
LMArena Chinese15281493
LMArena French14991461
LMArena German14991439
LMArena Japanese14531392
LMArena Korean14721395
LMArena Russian14631432
LMArena Spanish14601456

Instruction Following GLM-5.3 leads

GLM-5.3: 77.5 (#23), Hy3: 75.1 (#70)

BenchmarkGLM-5.3Hy3
LMArena Instruction Following14771426

Long Context GLM-5.3 leads

GLM-5.3: 45.4 (#41), Hy3: 44.1 (#75)

BenchmarkGLM-5.3Hy3
LMArena Longer Query14821442

Writing & Preference GLM-5.3 leads

GLM-5.3: 75.7 (#6), Hy3: 62.2 (#81)

BenchmarkGLM-5.3Hy3
LMArena Text14711439
LMArena Creative Writing14571402
LMArena Multi-Turn14721436
EQ-Bench Creative Writing2075—

Frequently asked questions

Is GLM-5.3 better than Hy3?

GLM-5.3 is the stronger model overall, scoring 54.8 to 44.2 on the Noometry Index. Hy3 costs 15× less per token, which makes it the better buy when GLM-5.3's lead doesn't matter for your workload.

Which is cheaper, GLM-5.3 or Hy3?

Hy3 is cheaper. It lists at $0.0825 per million input tokens and $0.33 per million output tokens; GLM-5.3 lists at $1.40 and $4.40.

Is GLM-5.3 or Hy3 better for coding?

GLM-5.3 scores higher on coding benchmarks: 59.5 versus 46.8 in the Noometry coding category.

Which has the bigger context window?

GLM-5.3 does, with 1M tokens against 262K.

How many benchmarks do GLM-5.3 and Hy3 share?

19 benchmarks have published results for both models. GLM-5.3 has 42 scored results on Noometry and Hy3 has 19.

Related comparisons

Go deeper