Model comparison

DeepSeek-V3.2-Exp vs GLM-4.5V

DeepSeek-V3.2-Exp is the stronger model overall, scoring 44.3 to 39.8 on the Noometry Index.

Last verified October 10, 2026. 14 shared benchmarks.

DeepSeek-V3.2-Exp

DeepSeek

44.3

Rank #78 Confirmed

GLM-4.5V Z.ai (Zhipu)

39.8

Rank #158 Confirmed

Summary

They share 14 benchmarks with published results for both. DeepSeek-V3.2-Exp scores higher in 7 categories and GLM-4.5V in 1 category; 8 gaps are clear of the uncertainty.
The widest gap is in knowledge, where DeepSeek-V3.2-Exp leads 51.7 to 37.5.
The biggest single-benchmark swing is Kagi LLM Benchmark: 52.2% for DeepSeek-V3.2-Exp and 59.8% for GLM-4.5V.
DeepSeek-V3.2-Exp is cheaper at $0.26 / $0.38 per million input/output tokens, against $0.60 / $1.80 for GLM-4.5V.
DeepSeek-V3.2-Exp accepts more context: 164K tokens versus 64K.

Side by side

DeepSeek-V3.2-Exp and GLM-4.5V specifications
	DeepSeek-V3.2-Exp	GLM-4.5V
Provider	DeepSeek	Z.ai (Zhipu)
Noometry Index	44.3	39.8
Released	2025-09-29	2025-08-11
Weights	Open	Open
Context window	164K	64K
Max output	66K	16K
Input $ / M tokens	$0.26	$0.60
Output $ / M tokens	$0.38	$1.80
Results tracked	49	15

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

Category by category

Coding DeepSeek-V3.2-Exp leads

DeepSeek-V3.2-Exp: 46.5 (#65), GLM-4.5V: 39.5 (#155)

Coding benchmarks
Benchmark	DeepSeek-V3.2-Exp	GLM-4.5V
LMArena Coding	1454	1347
SWE-bench Verified (bash only)	70%	—
Aider Polyglot	74.2%	—
LMArena WebDev	1362	—
SWE-bench Multilingual	59%	—
SciCode	38.9%	—
WeirdML	39.5%	—

Agentic & Tool Use Not comparable

DeepSeek-V3.2-Exp: 32.7 (#59), GLM-4.5V: —

Agentic & Tool Use benchmarks
Benchmark	DeepSeek-V3.2-Exp	GLM-4.5V
Terminal-Bench	39.6%	—
APEX-Agents	21.3%	—
Berkeley Function Calling Leaderboard	56.7%	—
TheAgentCompany	42.9%	—
Vending-Bench 2	1,034	—

Reasoning GLM-4.5V leads

DeepSeek-V3.2-Exp: 22.1 (#208), GLM-4.5V: 27.4 (#119)

Reasoning benchmarks
Benchmark	DeepSeek-V3.2-Exp	GLM-4.5V
Kagi LLM Benchmark	52.2%	59.8%
LMArena Hard Prompts	1434	1334
ARC-AGI-2	4%	—
NYT Connections (extended)	36.7%	—
ARC-AGI-1	57%	—
CritPt	2.9%	—
Chess Puzzles	14%	—
Thematic Generalization	65%	—
DTBench	87.7%	—
LMCA	29.1%	—
Epoch Capabilities Index	146.27	—

Math DeepSeek-V3.2-Exp leads

DeepSeek-V3.2-Exp: 41.7 (#87), GLM-4.5V: 37.4 (#159)

Math benchmarks
Benchmark	DeepSeek-V3.2-Exp	GLM-4.5V
LMArena Math	1435	1354
MathArena Final-Answer Competitions	57.7%	—
OTIS Mock AIME 2024-2025	87.8%	—
ProofBench	8%	—
FrontierMath (Feb 2025 set)	22.1%	—
FrontierMath Tier 4 (v1)	2.1%	—

Knowledge DeepSeek-V3.2-Exp leads

DeepSeek-V3.2-Exp: 51.7 (#66), GLM-4.5V: 37.5 (#156)

Knowledge benchmarks
Benchmark	DeepSeek-V3.2-Exp	GLM-4.5V
LMArena Expert	1436	1353
GPQA Diamond	83.4%	—
Vectara Hallucination Rate	5.3%	—

Multimodal Not comparable

DeepSeek-V3.2-Exp: —, GLM-4.5V: 34.3 (#92)

Multimodal benchmarks
Benchmark	DeepSeek-V3.2-Exp	GLM-4.5V
LMArena Vision	—	1154

Multilingual DeepSeek-V3.2-Exp leads

DeepSeek-V3.2-Exp: 52.2 (#90), GLM-4.5V: 44.6 (#177)

Multilingual benchmarks
Benchmark	DeepSeek-V3.2-Exp	GLM-4.5V
LMArena Non-English	1409	1303
LMArena Chinese	1461	1337
LMArena Russian	1424	1298
LMArena Spanish	1440	1336
LMArena French	1433	—
LMArena German	1440	—
LMArena Japanese	1374	—
LMArena Korean	1371	—

Instruction Following DeepSeek-V3.2-Exp leads

DeepSeek-V3.2-Exp: 74.5 (#93), GLM-4.5V: 69.2 (#175)

Instruction Following benchmarks
Benchmark	DeepSeek-V3.2-Exp	GLM-4.5V
LMArena Instruction Following	1413	1311

Long Context DeepSeek-V3.2-Exp leads

DeepSeek-V3.2-Exp: 47.6 (#16), GLM-4.5V: 39.6 (#171)

Long Context benchmarks
Benchmark	DeepSeek-V3.2-Exp	GLM-4.5V
LMArena Longer Query	1428	1304
Fiction.LiveBench	83.3%	—
CL-bench	13.2%	—
CL-bench Life	9.5%	—

Writing & Preference DeepSeek-V3.2-Exp leads

DeepSeek-V3.2-Exp: 62.4 (#77), GLM-4.5V: 52.5 (#170)

Writing & Preference benchmarks
Benchmark	DeepSeek-V3.2-Exp	GLM-4.5V
LMArena Text	1425	1333
LMArena Creative Writing	1403	1295
LMArena Multi-Turn	1427	1332
EQ-Bench Creative Writing	1515	—

Frequently asked questions

Is DeepSeek-V3.2-Exp better than GLM-4.5V?

DeepSeek-V3.2-Exp is the stronger model overall, scoring 44.3 to 39.8 on the Noometry Index.

Which is cheaper, DeepSeek-V3.2-Exp or GLM-4.5V?

DeepSeek-V3.2-Exp is cheaper. It lists at $0.26 per million input tokens and $0.38 per million output tokens; GLM-4.5V lists at $0.60 and $1.80.

Is DeepSeek-V3.2-Exp or GLM-4.5V better for coding?

DeepSeek-V3.2-Exp scores higher on coding benchmarks: 46.5 versus 39.5 in the Noometry coding category.

Which has the bigger context window?

DeepSeek-V3.2-Exp does, with 164K tokens against 64K.

How many benchmarks do DeepSeek-V3.2-Exp and GLM-4.5V share?

14 benchmarks have published results for both models. DeepSeek-V3.2-Exp has 49 scored results on Noometry and GLM-4.5V has 15.