Noometry

Model comparison

Claude 3 Haiku vs GLM-4.6

GLM-4.6 is the stronger model overall, scoring 41.4 to 25.9 on the Noometry Index.

Last verified . 19 shared benchmarks.

Claude 3 Haiku Anthropic

25.9

Rank #340 Confirmed

GLM-4.6 Z.ai (Zhipu)

41.4

Rank #135 Confirmed

Summary

  • They share 19 benchmarks with published results for both. Claude 3 Haiku scores higher in 0 categories and GLM-4.6 in 8 categories; 8 gaps are clear of the uncertainty.
  • The widest gap is in writing & preference, where GLM-4.6 leads 61.1 to 29.7.
  • The biggest single-benchmark swing is Kagi LLM Benchmark: 34.2% for Claude 3 Haiku and 47.4% for GLM-4.6.
  • GLM-4.6 has downloadable open weights; the other is API-only.

Side by side

Claude 3 HaikuGLM-4.6
ProviderAnthropicZ.ai (Zhipu)
Noometry Index25.941.4
Released2024-03-072025-09-30
WeightsProprietaryOpen
Context window—205K
Max output—131K
Input $ / M tokens—$0.60
Output $ / M tokens—$2.20
Results tracked3729

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

Category by category

Coding GLM-4.6 leads

Claude 3 Haiku: 26.4 (#325), GLM-4.6: 40.1 (#148)

BenchmarkClaude 3 HaikuGLM-4.6
LMArena Coding11991449
SWE-bench Verified (bash only)—55.4%
LMArena WebDev—1340
SciCode—38.4%
WeirdML9.8%—
BigCodeBench Instruct39.4%—
BigCodeBench Complete50.1%—
CadEval12%—
ALE-Bench—340.82
HumanEval+68.9%—
MBPP+68.8%—

Agentic & Tool Use Not comparable

Claude 3 Haiku: —, GLM-4.6: 32.3 (#66)

BenchmarkClaude 3 HaikuGLM-4.6
Terminal-Bench—24.5%
Berkeley Function Calling Leaderboard—72.4%

Reasoning GLM-4.6 leads

Claude 3 Haiku: 16.3 (#307), GLM-4.6: 23.7 (#172)

BenchmarkClaude 3 HaikuGLM-4.6
Kagi LLM Benchmark34.2%47.4%
LMArena Hard Prompts11741440
CritPt—1.1%
DTBench50.1%—
LMCA8.8%—
Epoch Capabilities Index118.35—
ForecastBench53.2—
WinoGrande74.2%—

Math GLM-4.6 leads

Claude 3 Haiku: 9.8 (#319), GLM-4.6: 39.1 (#111)

BenchmarkClaude 3 HaikuGLM-4.6
LMArena Math11881432
OTIS Mock AIME 2024-20251.8%—
MATH Level 514.9%—
FrontierMath (Feb 2025 set)—3.8%
FrontierMath Tier 4 (v1)—2.1%

Knowledge GLM-4.6 leads

Claude 3 Haiku: 17.3 (#285), GLM-4.6: 40.2 (#124)

BenchmarkClaude 3 HaikuGLM-4.6
LMArena Expert11481431
GPQA Diamond36.3%—
Confabulations34.2%—
Vectara Hallucination Rate—9.5%
MMLU73.8%—

Multimodal Not comparable

Claude 3 Haiku: 23.6 (#128), GLM-4.6: —

BenchmarkClaude 3 HaikuGLM-4.6
LMArena Vision950—
ScienceQA72%—

Multilingual GLM-4.6 leads

Claude 3 Haiku: 36.0 (#243), GLM-4.6: 53.5 (#66)

BenchmarkClaude 3 HaikuGLM-4.6
LMArena Non-English11781426
LMArena Chinese11551499
LMArena French11951459
LMArena German11741447
LMArena Japanese11021393
LMArena Korean11091400
LMArena Russian12041419
LMArena Spanish11661436

Instruction Following GLM-4.6 leads

Claude 3 Haiku: 61.3 (#247), GLM-4.6: 74.3 (#98)

BenchmarkClaude 3 HaikuGLM-4.6
LMArena Instruction Following11731410

Long Context GLM-4.6 leads

Claude 3 Haiku: 36.1 (#237), GLM-4.6: 43.4 (#94)

BenchmarkClaude 3 HaikuGLM-4.6
LMArena Longer Query11901422

Writing & Preference GLM-4.6 leads

Claude 3 Haiku: 29.7 (#291), GLM-4.6: 61.1 (#90)

BenchmarkClaude 3 HaikuGLM-4.6
LMArena Text11951440
LMArena Creative Writing11571411
EQ-Bench Creative Writing7171411
LMArena Multi-Turn11901427

Frequently asked questions

Is Claude 3 Haiku better than GLM-4.6?

GLM-4.6 is the stronger model overall, scoring 41.4 to 25.9 on the Noometry Index.

Is Claude 3 Haiku or GLM-4.6 better for coding?

GLM-4.6 scores higher on coding benchmarks: 40.1 versus 26.4 in the Noometry coding category.

How many benchmarks do Claude 3 Haiku and GLM-4.6 share?

19 benchmarks have published results for both models. Claude 3 Haiku has 37 scored results on Noometry and GLM-4.6 has 29.

Related comparisons

Go deeper