Noometry

Model comparison

Claude 3 Haiku vs GLM-5.1

GLM-5.1 is the stronger model overall, scoring 47.8 to 25.9 on the Noometry Index.

Last verified . 22 shared benchmarks.

Claude 3 Haiku Anthropic

25.9

Rank #340 Confirmed

GLM-5.1 Z.ai (Zhipu)

47.8

Rank #59 Confirmed

Summary

  • They share 22 benchmarks with published results for both. Claude 3 Haiku scores higher in 0 categories and GLM-5.1 in 8 categories; 8 gaps are clear of the uncertainty.
  • The widest gap is in math, where GLM-5.1 leads 49.7 to 9.8.
  • The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 1.8% for Claude 3 Haiku and 93.3% for GLM-5.1.
  • GLM-5.1 has downloadable open weights; the other is API-only.

Side by side

Claude 3 HaikuGLM-5.1
ProviderAnthropicZ.ai (Zhipu)
Noometry Index25.947.8
Released2024-03-072026-04-07
WeightsProprietaryOpen
Context window—200K
Max output—131K
Input $ / M tokens—$1.40
Output $ / M tokens—$4.40
Results tracked3741

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

Category by category

Coding GLM-5.1 leads

Claude 3 Haiku: 26.4 (#325), GLM-5.1: 48.7 (#55)

BenchmarkClaude 3 HaikuGLM-5.1
WeirdML9.8%57.1%
LMArena Coding11991485
SWE-bench Verified—74.2%
LMArena WebDev—1508
SciCode—43.8%
BigCodeBench Instruct39.4%—
BigCodeBench Complete50.1%—
CadEval12%—
ALE-Bench—887.1
HumanEval+68.9%—
MBPP+68.8%—

Agentic & Tool Use Not comparable

Claude 3 Haiku: —, GLM-5.1: 24.9 (#113)

BenchmarkClaude 3 HaikuGLM-5.1
APEX-Agents—40.9%
ExploitBench—18.1%
GBAEval—0%
Vending-Bench 2—5,634

Reasoning GLM-5.1 leads

Claude 3 Haiku: 16.3 (#307), GLM-5.1: 39.1 (#60)

BenchmarkClaude 3 HaikuGLM-5.1
LMArena Hard Prompts11741472
Epoch Capabilities Index118.35149.84
SimpleBench—55.1%
Kagi LLM Benchmark34.2%—
NYT Connections (extended)—77.7%
CritPt—4.6%
Chess Puzzles—19%
Thematic Generalization—69.8%
DTBench50.1%—
LMCA8.8%—
ForecastBench53.2—
WinoGrande74.2%—

Math GLM-5.1 leads

Claude 3 Haiku: 9.8 (#319), GLM-5.1: 49.7 (#60)

BenchmarkClaude 3 HaikuGLM-5.1
OTIS Mock AIME 2024-20251.8%93.3%
LMArena Math11881473
FrontierMath (Tiers 1-3)—36.8%
MathArena Final-Answer Competitions—67.1%
ProofBench—22.2%
MATH Level 514.9%—
FrontierMath (Feb 2025 set)—33.4%
FrontierMath Tier 4 (v1)—12.5%

Knowledge GLM-5.1 leads

Claude 3 Haiku: 17.3 (#285), GLM-5.1: 54.9 (#50)

BenchmarkClaude 3 HaikuGLM-5.1
GPQA Diamond36.3%89.9%
LMArena Expert11481476
SimpleQA Verified—34%
Confabulations34.2%—
MMLU73.8%—

Multimodal Not comparable

Claude 3 Haiku: 23.6 (#128), GLM-5.1: —

BenchmarkClaude 3 HaikuGLM-5.1
LMArena Vision950—
ScienceQA72%—

Multilingual GLM-5.1 leads

Claude 3 Haiku: 36.0 (#243), GLM-5.1: 55.0 (#36)

BenchmarkClaude 3 HaikuGLM-5.1
LMArena Non-English11781447
LMArena Chinese11551515
LMArena French11951474
LMArena German11741465
LMArena Japanese11021434
LMArena Korean11091418
LMArena Russian12041454
LMArena Spanish11661469

Instruction Following GLM-5.1 leads

Claude 3 Haiku: 61.3 (#247), GLM-5.1: 76.3 (#42)

BenchmarkClaude 3 HaikuGLM-5.1
LMArena Instruction Following11731451

Long Context GLM-5.1 leads

Claude 3 Haiku: 36.1 (#237), GLM-5.1: 44.9 (#53)

BenchmarkClaude 3 HaikuGLM-5.1
LMArena Longer Query11901466

Writing & Preference GLM-5.1 leads

Claude 3 Haiku: 29.7 (#291), GLM-5.1: 66.9 (#31)

BenchmarkClaude 3 HaikuGLM-5.1
LMArena Text11951461
LMArena Creative Writing11571453
EQ-Bench Creative Writing7171592
LMArena Multi-Turn11901472

Frequently asked questions

Is Claude 3 Haiku better than GLM-5.1?

GLM-5.1 is the stronger model overall, scoring 47.8 to 25.9 on the Noometry Index.

Is Claude 3 Haiku or GLM-5.1 better for coding?

GLM-5.1 scores higher on coding benchmarks: 48.7 versus 26.4 in the Noometry coding category.

How many benchmarks do Claude 3 Haiku and GLM-5.1 share?

22 benchmarks have published results for both models. Claude 3 Haiku has 37 scored results on Noometry and GLM-5.1 has 41.

Related comparisons

Go deeper