Noometry

Model comparison

GLM-5.3-Flash vs Hy4 preview

GLM-5.3-Flash is the stronger model overall, scoring 49.6 to 44.6 on the Noometry Index.

Last verified . 2 shared benchmarks.

GLM-5.3-Flash Z.ai (Zhipu)

49.6

Rank #45 Confirmed

Hy4 preview Tencent

44.6

Rank #73 Reported

Summary

  • They share 2 benchmarks with published results for both. GLM-5.3-Flash scores higher in 2 categories and Hy4 preview in 1 category; 3 gaps are clear of the uncertainty.
  • The biggest single-benchmark swing is ProofBench: 21% for GLM-5.3-Flash and 75% for Hy4 preview.
  • GLM-5.3-Flash is cheaper at $0.15 / $0.50 per million input/output tokens, against $0.83 / $2.50 for Hy4 preview.
  • Hy4 preview accepts more context: 1.05M tokens versus 1M.

Side by side

GLM-5.3-FlashHy4 preview
ProviderZ.ai (Zhipu)Tencent
Noometry Index49.644.6
Released2026-08-202026-08-28
WeightsOpenOpen
Context window1M1.05M
Max output131K64K
Input $ / M tokens$0.15$0.83
Output $ / M tokens$0.50$2.50
Results tracked363

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: 53.7 (#26), Hy4 preview: 51.4 (#36)

BenchmarkGLM-5.3-FlashHy4 preview
LMArena WebDev16141633
DeepSWE63.4%—
FrontierCode31.8%—
SciCode51.6%—
LMArena Coding1506—
ALE-Bench303.55—

Agentic & Tool Use Not comparable

GLM-5.3-Flash: 34.3 (#41), Hy4 preview: —

BenchmarkGLM-5.3-FlashHy4 preview
APEX-Agents52.8%—
GDP.pdf14%—

Reasoning GLM-5.3-Flash leads

GLM-5.3-Flash: 31.7 (#69), Hy4 preview: 30.6 (#75)

BenchmarkGLM-5.3-FlashHy4 preview
NYT Connections (extended)—68.2%
CritPt15.4%—
Chess Puzzles14%—
LMArena Hard Prompts1490—
Mystery Game Puzzles8%—
Surface Evolver Bench52.5%—
Bench to the Future 30.15—
Epoch Capabilities Index151.9—

Math Hy4 preview leads

GLM-5.3-Flash: 53.1 (#38), Hy4 preview: 54.5 (#37)

BenchmarkGLM-5.3-FlashHy4 preview
ProofBench21%75%
FrontierMath (Tiers 1-3)55.8%—
FrontierMath Tier 417.1%—
OTIS Mock AIME 2024-202593.9%—
LMArena Math1499—

Knowledge Not comparable

GLM-5.3-Flash: 58.4 (#33), Hy4 preview: —

BenchmarkGLM-5.3-FlashHy4 preview
GPQA Diamond90.2%—
LMArena Expert1512—

Multimodal Not comparable

GLM-5.3-Flash: 42.6 (#25), Hy4 preview: —

BenchmarkGLM-5.3-FlashHy4 preview
LMArena Vision1304—

Multilingual Not comparable

GLM-5.3-Flash: 56.2 (#24), Hy4 preview: —

BenchmarkGLM-5.3-FlashHy4 preview
LMArena Non-English1462—
LMArena Chinese1529—
LMArena French1495—
LMArena German1457—
LMArena Japanese1420—
LMArena Korean1443—
LMArena Russian1472—
LMArena Spanish1471—

Instruction Following Not comparable

GLM-5.3-Flash: 77.8 (#17), Hy4 preview: —

BenchmarkGLM-5.3-FlashHy4 preview
LMArena Instruction Following1480—

Long Context Not comparable

GLM-5.3-Flash: 45.3 (#41), Hy4 preview: —

BenchmarkGLM-5.3-FlashHy4 preview
LMArena Longer Query1479—

Writing & Preference Not comparable

GLM-5.3-Flash: 65.4 (#49), Hy4 preview: —

BenchmarkGLM-5.3-FlashHy4 preview
LMArena Text1471—
LMArena Creative Writing1440—
LMArena Multi-Turn1471—

Frequently asked questions

Is GLM-5.3-Flash better than Hy4 preview?

GLM-5.3-Flash is the stronger model overall, scoring 49.6 to 44.6 on the Noometry Index.

Which is cheaper, GLM-5.3-Flash or Hy4 preview?

GLM-5.3-Flash is cheaper. It lists at $0.15 per million input tokens and $0.50 per million output tokens; Hy4 preview lists at $0.83 and $2.50.

Is GLM-5.3-Flash or Hy4 preview better for coding?

GLM-5.3-Flash scores higher on coding benchmarks: 53.7 versus 51.4 in the Noometry coding category.

Which has the bigger context window?

Hy4 preview does, with 1.05M tokens against 1M.

How many benchmarks do GLM-5.3-Flash and Hy4 preview share?

2 benchmarks have published results for both models. GLM-5.3-Flash has 36 scored results on Noometry and Hy4 preview has 3.

Related comparisons

Go deeper