Noometry

Model comparison

Claude 3 Haiku vs GLM-5.2

GLM-5.2 is the stronger model overall, scoring 51.1 to 25.9 on the Noometry Index.

Last verified . 25 shared benchmarks.

Claude 3 Haiku Anthropic

25.9

Rank #340 Confirmed

GLM-5.2 Z.ai (Zhipu)

51.1

Rank #44 Confirmed

Summary

  • They share 25 benchmarks with published results for both. Claude 3 Haiku scores higher in 0 categories and GLM-5.2 in 8 categories; 8 gaps are clear of the uncertainty.
  • The widest gap is in math, where GLM-5.2 leads 55.7 to 9.8.
  • The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 1.8% for Claude 3 Haiku and 86.4% for GLM-5.2.
  • GLM-5.2 has downloadable open weights; the other is API-only.

Side by side

Claude 3 HaikuGLM-5.2
ProviderAnthropicZ.ai (Zhipu)
Noometry Index25.951.1
Released2024-03-072026-06-13
WeightsProprietaryOpen
Context window—1M
Max output—131K
Input $ / M tokens—$1.40
Output $ / M tokens—$4.40
Results tracked3751

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

Category by category

Coding GLM-5.2 leads

Claude 3 Haiku: 26.4 (#325), GLM-5.2: 51.3 (#41)

BenchmarkClaude 3 HaikuGLM-5.2
WeirdML9.8%70.1%
LMArena Coding11991485
SWE-bench Verified—78.7%
DeepSWE—43.8%
FrontierCode—24.5%
LMArena WebDev—1603
SciCode—50.5%
BigCodeBench Instruct39.4%—
BigCodeBench Complete50.1%—
CadEval12%—
ALE-Bench—1,047
HumanEval+68.9%—
MBPP+68.8%—

Agentic & Tool Use Not comparable

Claude 3 Haiku: —, GLM-5.2: 32.4 (#63)

BenchmarkClaude 3 HaikuGLM-5.2
APEX-Agents—45.2%
τ²-bench Banking—37.1%
PostTrainBench—31.7%
GBAEval—0%
Vending-Bench 2—8,314

Reasoning GLM-5.2 leads

Claude 3 Haiku: 16.3 (#307), GLM-5.2: 42.3 (#52)

BenchmarkClaude 3 HaikuGLM-5.2
Kagi LLM Benchmark34.2%62.6%
LMArena Hard Prompts11741480
DTBench50.1%93.6%
LMCA8.8%45.8%
Epoch Capabilities Index118.35151.78
ARC-AGI-2—22.8%
SimpleBench—58.8%
NYT Connections (extended)—74.3%
ARC-AGI-1—77%
CritPt—20.9%
Chess Puzzles—21%
EBR-Bench—9.5%
Mystery Game Puzzles—19%
Surface Evolver Bench—55.6%
ForecastBench53.2—
WinoGrande74.2%—

Math GLM-5.2 leads

Claude 3 Haiku: 9.8 (#319), GLM-5.2: 55.7 (#43)

BenchmarkClaude 3 HaikuGLM-5.2
OTIS Mock AIME 2024-20251.8%86.4%
LMArena Math11881482
FrontierMath (Tiers 1-3)—59.2%
FrontierMath Tier 4—29.3%
MathArena Final-Answer Competitions—67.6%
ProofBench—35%
MATH Level 514.9%—

Knowledge GLM-5.2 leads

Claude 3 Haiku: 17.3 (#285), GLM-5.2: 57.1 (#40)

BenchmarkClaude 3 HaikuGLM-5.2
GPQA Diamond36.3%91.9%
LMArena Expert11481486
SimpleQA Verified—34.2%
Confabulations34.2%—
MMLU73.8%—

Multimodal Not comparable

Claude 3 Haiku: 23.6 (#128), GLM-5.2: —

BenchmarkClaude 3 HaikuGLM-5.2
LMArena Vision950—
ScienceQA72%—

Multilingual GLM-5.2 leads

Claude 3 Haiku: 36.0 (#243), GLM-5.2: 55.8 (#26)

BenchmarkClaude 3 HaikuGLM-5.2
LMArena Non-English11781459
LMArena Chinese11551519
LMArena French11951479
LMArena German11741468
LMArena Japanese11021451
LMArena Korean11091445
LMArena Russian12041466
LMArena Spanish11661477

Instruction Following GLM-5.2 leads

Claude 3 Haiku: 61.3 (#247), GLM-5.2: 76.9 (#34)

BenchmarkClaude 3 HaikuGLM-5.2
LMArena Instruction Following11731465

Long Context GLM-5.2 leads

Claude 3 Haiku: 36.1 (#237), GLM-5.2: 45.3 (#43)

BenchmarkClaude 3 HaikuGLM-5.2
LMArena Longer Query11901479

Writing & Preference GLM-5.2 leads

Claude 3 Haiku: 29.7 (#291), GLM-5.2: 70.4 (#21)

BenchmarkClaude 3 HaikuGLM-5.2
LMArena Text11951470
LMArena Creative Writing11571462
EQ-Bench Creative Writing7171757
LMArena Multi-Turn11901469
EQ-Bench 4—1222

Frequently asked questions

Is Claude 3 Haiku better than GLM-5.2?

GLM-5.2 is the stronger model overall, scoring 51.1 to 25.9 on the Noometry Index.

Is Claude 3 Haiku or GLM-5.2 better for coding?

GLM-5.2 scores higher on coding benchmarks: 51.3 versus 26.4 in the Noometry coding category.

How many benchmarks do Claude 3 Haiku and GLM-5.2 share?

25 benchmarks have published results for both models. Claude 3 Haiku has 37 scored results on Noometry and GLM-5.2 has 51.

Related comparisons

Go deeper