Noometry

Model comparison

GLM-4.5 vs Llama 3-8B

GLM-4.5 is the stronger model overall, scoring 41.4 to 25.3 on the Noometry Index. Llama 3-8B costs 7.1× less per token, which makes it the better buy when GLM-4.5's lead doesn't matter for your workload.

Last verified . 17 shared benchmarks.

GLM-4.5 Z.ai (Zhipu)

41.4

Rank #125 Confirmed

Llama 3-8B Meta

25.3

Rank #338 Confirmed

Summary

  • They share 17 benchmarks with published results for both. GLM-4.5 scores higher in 8 categories and Llama 3-8B in 0 categories; 8 gaps are clear of the uncertainty.
  • The widest gap is in math, where GLM-4.5 leads 36.3 to 8.3.
  • Llama 3-8B is cheaper at $0.14 / $0.14 per million input/output tokens, against $0.60 / $2.20 for GLM-4.5.
  • GLM-4.5 accepts more context: 131K tokens versus 8K.

Side by side

GLM-4.5Llama 3-8B
ProviderZ.ai (Zhipu)Meta
Noometry Index41.425.3
Released2025-07-272024-04-18
WeightsOpenOpen
Context window131K8K
Max output98K8K
Input $ / M tokens$0.60$0.14
Output $ / M tokens$2.20$0.14
Results tracked2734

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

Category by category

Coding GLM-4.5 leads

GLM-4.5: 41.1 (#121), Llama 3-8B: 30.5 (#285)

BenchmarkGLM-4.5Llama 3-8B
LMArena Coding14341152
SWE-bench Verified (bash only)54.2%—
WeirdML40.6%—
BigCodeBench Instruct—31.9%
BigCodeBench Complete—36.9%
ALE-Bench344.82—
AlgoTune1.52—
HumanEval+—56.7%
MBPP+—54.8%

Reasoning GLM-4.5 leads

GLM-4.5: 27.7 (#99), Llama 3-8B: 13.8 (#321)

BenchmarkGLM-4.5Llama 3-8B
LMArena Hard Prompts14301133
Kagi LLM Benchmark57.9%—
Chess Puzzles—0%
DTBench—43.9%
Adversarial NLI—57.3%
Epoch Capabilities Index—116.33
ForecastBench—58.6
WinoGrande—75.7%

Math GLM-4.5 leads

GLM-4.5: 36.3 (#119), Llama 3-8B: 8.3 (#317)

BenchmarkGLM-4.5Llama 3-8B
LMArena Math14271151
OTIS Mock AIME 2024-2025—1.9%
MATH Level 5—6.1%

Knowledge GLM-4.5 leads

GLM-4.5: 35.5 (#172), Llama 3-8B: 7.8 (#303)

BenchmarkGLM-4.5Llama 3-8B
LMArena Expert14331113
GPQA Diamond—26.1%
Humanity's Last Exam8.3%—
Confabulations11.3%—
ARC (AI2) Challenge—82.8%
MMLU—68.8%
OpenBookQA—82.6%
TriviaQA—67.7%

Multilingual GLM-4.5 leads

GLM-4.5: 53.0 (#74), Llama 3-8B: 31.0 (#256)

BenchmarkGLM-4.5Llama 3-8B
LMArena Non-English14171098
LMArena Chinese14651076
LMArena French14231158
LMArena German14061104
LMArena Japanese1416967
LMArena Korean13821003
LMArena Russian14141109
LMArena Spanish14541172

Instruction Following GLM-4.5 leads

GLM-4.5: 74.3 (#100), Llama 3-8B: 58.7 (#255)

BenchmarkGLM-4.5Llama 3-8B
LMArena Instruction Following14051127

Long Context GLM-4.5 leads

GLM-4.5: 38.2 (#196), Llama 3-8B: 34.2 (#246)

BenchmarkGLM-4.5Llama 3-8B
LMArena Longer Query14131128
Fiction.LiveBench58.3%—

Writing & Preference GLM-4.5 leads

GLM-4.5: 57.7 (#121), Llama 3-8B: 37.7 (#251)

BenchmarkGLM-4.5Llama 3-8B
LMArena Text14301166
LMArena Creative Writing13951150
LMArena Multi-Turn14151152
Short-Story Creative Writing73.4%—
EQ-Bench Creative Writing1343—

Frequently asked questions

Is GLM-4.5 better than Llama 3-8B?

GLM-4.5 is the stronger model overall, scoring 41.4 to 25.3 on the Noometry Index. Llama 3-8B costs 7.1× less per token, which makes it the better buy when GLM-4.5's lead doesn't matter for your workload.

Which is cheaper, GLM-4.5 or Llama 3-8B?

Llama 3-8B is cheaper. It lists at $0.14 per million input tokens and $0.14 per million output tokens; GLM-4.5 lists at $0.60 and $2.20.

Is GLM-4.5 or Llama 3-8B better for coding?

GLM-4.5 scores higher on coding benchmarks: 41.1 versus 30.5 in the Noometry coding category.

Which has the bigger context window?

GLM-4.5 does, with 131K tokens against 8K.

How many benchmarks do GLM-4.5 and Llama 3-8B share?

17 benchmarks have published results for both models. GLM-4.5 has 27 scored results on Noometry and Llama 3-8B has 34.

Related comparisons

Go deeper