Noometry

Model comparison

Claude 3 Haiku vs GLM-5.3

GLM-5.3 is the stronger model overall, scoring 54.8 to 25.9 on the Noometry Index.

Last verified . 24 shared benchmarks.

Claude 3 Haiku Anthropic

25.9

Rank #340 Confirmed

GLM-5.3 Z.ai (Zhipu)

54.8

Rank #26 Confirmed

Summary

  • They share 24 benchmarks with published results for both. Claude 3 Haiku scores higher in 0 categories and GLM-5.3 in 8 categories; 8 gaps are clear of the uncertainty.
  • The widest gap is in math, where GLM-5.3 leads 62.3 to 9.8.
  • The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 1.8% for Claude 3 Haiku and 91.1% for GLM-5.3.
  • GLM-5.3 has downloadable open weights; the other is API-only.

Side by side

Claude 3 HaikuGLM-5.3
ProviderAnthropicZ.ai (Zhipu)
Noometry Index25.954.8
Released2024-03-072026-08-14
WeightsProprietaryOpen
Context window—1M
Max output—131K
Input $ / M tokens—$1.40
Output $ / M tokens—$4.40
Results tracked3742

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

Category by category

Coding GLM-5.3 leads

Claude 3 Haiku: 26.4 (#325), GLM-5.3: 59.5 (#14)

BenchmarkClaude 3 HaikuGLM-5.3
WeirdML9.8%75.4%
LMArena Coding11991496
DeepSWE—69%
FrontierCode—40.1%
CursorBench—42.6%
LMArena WebDev—1622
FrontierSWE—30.2%
SciCode—59%
BigCodeBench Instruct39.4%—
BigCodeBench Complete50.1%—
CadEval12%—
ALE-Bench—1,317
HumanEval+68.9%—
MBPP+68.8%—

Agentic & Tool Use Not comparable

Claude 3 Haiku: —, GLM-5.3: 36.4 (#38)

BenchmarkClaude 3 HaikuGLM-5.3
APEX-Agents—56.6%
Vending-Bench 2—8,164

Reasoning GLM-5.3 leads

Claude 3 Haiku: 16.3 (#307), GLM-5.3: 46.1 (#46)

BenchmarkClaude 3 HaikuGLM-5.3
LMArena Hard Prompts11741489
DTBench50.1%87.7%
LMCA8.8%55.5%
Epoch Capabilities Index118.35155.61
Kagi LLM Benchmark34.2%—
NYT Connections (extended)—74.2%
CritPt—19.1%
Chess Puzzles—21%
Mystery Game Puzzles—33%
Bench to the Future 3—0.15
ForecastBench53.2—
WinoGrande74.2%—

Math GLM-5.3 leads

Claude 3 Haiku: 9.8 (#319), GLM-5.3: 62.3 (#33)

BenchmarkClaude 3 HaikuGLM-5.3
OTIS Mock AIME 2024-20251.8%91.1%
LMArena Math11881489
FrontierMath (Tiers 1-3)—68.8%
FrontierMath Tier 4—29.3%
ProofBench—49%
MATH Level 514.9%—

Knowledge GLM-5.3 leads

Claude 3 Haiku: 17.3 (#285), GLM-5.3: 58.3 (#37)

BenchmarkClaude 3 HaikuGLM-5.3
GPQA Diamond36.3%90.9%
LMArena Expert11481516
SimpleQA Verified—41%
Confabulations34.2%—
MMLU73.8%—

Multimodal Not comparable

Claude 3 Haiku: 23.6 (#128), GLM-5.3: —

BenchmarkClaude 3 HaikuGLM-5.3
LMArena Vision950—
ScienceQA72%—

Multilingual GLM-5.3 leads

Claude 3 Haiku: 36.0 (#243), GLM-5.3: 55.7 (#28)

BenchmarkClaude 3 HaikuGLM-5.3
LMArena Non-English11781457
LMArena Chinese11551528
LMArena French11951499
LMArena German11741499
LMArena Japanese11021453
LMArena Korean11091472
LMArena Russian12041463
LMArena Spanish11661460

Instruction Following GLM-5.3 leads

Claude 3 Haiku: 61.3 (#247), GLM-5.3: 77.5 (#23)

BenchmarkClaude 3 HaikuGLM-5.3
LMArena Instruction Following11731477

Long Context GLM-5.3 leads

Claude 3 Haiku: 36.1 (#237), GLM-5.3: 45.4 (#41)

BenchmarkClaude 3 HaikuGLM-5.3
LMArena Longer Query11901482

Writing & Preference GLM-5.3 leads

Claude 3 Haiku: 29.7 (#291), GLM-5.3: 75.7 (#6)

BenchmarkClaude 3 HaikuGLM-5.3
LMArena Text11951471
LMArena Creative Writing11571457
EQ-Bench Creative Writing7172075
LMArena Multi-Turn11901472

Frequently asked questions

Is Claude 3 Haiku better than GLM-5.3?

GLM-5.3 is the stronger model overall, scoring 54.8 to 25.9 on the Noometry Index.

Is Claude 3 Haiku or GLM-5.3 better for coding?

GLM-5.3 scores higher on coding benchmarks: 59.5 versus 26.4 in the Noometry coding category.

How many benchmarks do Claude 3 Haiku and GLM-5.3 share?

24 benchmarks have published results for both models. Claude 3 Haiku has 37 scored results on Noometry and GLM-5.3 has 42.

Related comparisons

Go deeper