Noometry

Model comparison

GLM-4.5V vs Phi-4

GLM-4.5V is the stronger model overall, scoring 39.2 to 30.9 on the Noometry Index. Phi-4 costs 10× less per token, which makes it the better buy when GLM-4.5V's lead doesn't matter for your workload.

Last verified . 13 shared benchmarks.

GLM-4.5V Z.ai (Zhipu)

39.2

Rank #158 Confirmed

Phi-4 Microsoft

30.9

Rank #269 Confirmed

Summary

  • They share 13 benchmarks with published results for both. GLM-4.5V scores higher in 8 categories and Phi-4 in 0 categories; 8 gaps are clear of the uncertainty.
  • The widest gap is in math, where GLM-4.5V leads 34.7 to 19.6.
  • Phi-4 is cheaper at $0.07 / $0.14 per million input/output tokens, against $0.60 / $1.80 for GLM-4.5V.
  • Phi-4 accepts more context: 128K tokens versus 64K.

Side by side

GLM-4.5VPhi-4
ProviderZ.ai (Zhipu)Microsoft
Noometry Index39.230.9
Released2025-08-112024-12-11
WeightsOpenOpen
Context window64K128K
Max output16K4K
Input $ / M tokens$0.60$0.07
Output $ / M tokens$1.80$0.14
Results tracked1537

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

Category by category

Coding GLM-4.5V leads

GLM-4.5V: 38.9 (#152), Phi-4: 33.8 (#237)

BenchmarkGLM-4.5VPhi-4
LMArena Coding13471231
BigCodeBench Instruct—45.5%
LiveBench Coding—30.7%
BigCodeBench Complete—55.4%

Agentic & Tool Use Not comparable

GLM-4.5V: —, Phi-4: 22.6 (#120)

BenchmarkGLM-4.5VPhi-4
Berkeley Function Calling Leaderboard—28.8%
BALROG—11.6%

Reasoning GLM-4.5V leads

GLM-4.5V: 26.5 (#118), Phi-4: 17.0 (#287)

BenchmarkGLM-4.5VPhi-4
LMArena Hard Prompts13351220
Kagi LLM Benchmark59.8%—
Chess Puzzles—1%
LiveBench Reasoning—47.8%
LiveBench Data Analysis—45.2%
Epoch Capabilities Index—130.43
LiveBench—41.6%

Math GLM-4.5V leads

GLM-4.5V: 34.7 (#157), Phi-4: 19.6 (#279)

BenchmarkGLM-4.5VPhi-4
LMArena Math13531246
OTIS Mock AIME 2024-2025—13.8%
LiveBench Math—42%
MATH Level 5—64.9%

Knowledge GLM-4.5V leads

GLM-4.5V: 36.8 (#155), Phi-4: 32.5 (#204)

BenchmarkGLM-4.5VPhi-4
LMArena Expert13531203
GPQA Diamond—56.1%
Confabulations—29.4%
Vectara Hallucination Rate—3.7%
MMLU—84.8%

Multimodal Not comparable

GLM-4.5V: 33.6 (#86), Phi-4: —

BenchmarkGLM-4.5VPhi-4
LMArena Vision1154—

Multilingual GLM-4.5V leads

GLM-4.5V: 44.9 (#172), Phi-4: 37.4 (#232)

BenchmarkGLM-4.5VPhi-4
LMArena Non-English13041197
LMArena Chinese13381211
LMArena Russian12991209
LMArena Spanish13351234
LMArena French—1223
LMArena German—1222
LMArena Japanese—1158
LMArena Korean—1151

Instruction Following GLM-4.5V leads

GLM-4.5V: 69.4 (#170), Phi-4: 60.6 (#246)

BenchmarkGLM-4.5VPhi-4
LMArena Instruction Following13111201
LiveBench Instruction Following—58.4%

Long Context GLM-4.5V leads

GLM-4.5V: 39.6 (#166), Phi-4: 36.9 (#221)

BenchmarkGLM-4.5VPhi-4
LMArena Longer Query13051217

Writing & Preference GLM-4.5V leads

GLM-4.5V: 52.7 (#162), Phi-4: 40.7 (#239)

BenchmarkGLM-4.5VPhi-4
LMArena Text13331217
LMArena Creative Writing12951182
LMArena Multi-Turn13331206
Short-Story Creative Writing—62.6%
LiveBench Language—25.6%

Frequently asked questions

Is GLM-4.5V better than Phi-4?

GLM-4.5V is the stronger model overall, scoring 39.2 to 30.9 on the Noometry Index. Phi-4 costs 10× less per token, which makes it the better buy when GLM-4.5V's lead doesn't matter for your workload.

Which is cheaper, GLM-4.5V or Phi-4?

Phi-4 is cheaper. It lists at $0.07 per million input tokens and $0.14 per million output tokens; GLM-4.5V lists at $0.60 and $1.80.

Is GLM-4.5V or Phi-4 better for coding?

GLM-4.5V scores higher on coding benchmarks: 38.9 versus 33.8 in the Noometry coding category.

Which has the bigger context window?

Phi-4 does, with 128K tokens against 64K.

How many benchmarks do GLM-4.5V and Phi-4 share?

13 benchmarks have published results for both models. GLM-4.5V has 15 scored results on Noometry and Phi-4 has 37.

Related comparisons

Go deeper