Model comparison

DeepSeek-V3.1 vs GLM-4.5-Air

DeepSeek-V3.1 is the stronger model overall, scoring 42.8 to 38.9 on the Noometry Index.

Last verified October 10, 2026. 20 shared benchmarks.

DeepSeek-V3.1

DeepSeek

42.8

Rank #108 Confirmed

GLM-4.5-Air Z.ai (Zhipu)

38.9

Rank #177 Confirmed

Summary

They share 20 benchmarks with published results for both. DeepSeek-V3.1 scores higher in 7 categories and GLM-4.5-Air in 1 category; 8 gaps are clear of the uncertainty.
The widest gap is in knowledge, where DeepSeek-V3.1 leads 43.7 to 35.0.
The biggest single-benchmark swing is Kagi LLM Benchmark: 53.2% for DeepSeek-V3.1 and 43% for GLM-4.5-Air.
Both cost about the same: $0.25 input and $0.95 output per million tokens.
DeepSeek-V3.1 accepts more context: 164K tokens versus 131K.

Side by side

DeepSeek-V3.1 and GLM-4.5-Air specifications
	DeepSeek-V3.1	GLM-4.5-Air
Provider	DeepSeek	Z.ai (Zhipu)
Noometry Index	42.8	38.9
Released	2025-08-21	2025-07-20
Weights	Open	Open
Context window	164K	131K
Max output	8K	98K
Input $ / M tokens	$0.25	$0.20
Output $ / M tokens	$0.95	$1.10
Results tracked	27	27

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

Category by category

Coding DeepSeek-V3.1 leads

DeepSeek-V3.1: 40.3 (#144), GLM-4.5-Air: 33.3 (#259)

Coding benchmarks
Benchmark	DeepSeek-V3.1	GLM-4.5-Air
LMArena Coding	1417	1397
GSO	—	2.9%
WeirdML	38.4%	—

Reasoning DeepSeek-V3.1 leads

DeepSeek-V3.1: 27.9 (#110), GLM-4.5-Air: 24.1 (#166)

Reasoning benchmarks
Benchmark	DeepSeek-V3.1	GLM-4.5-Air
Kagi LLM Benchmark	53.2%	43%
LMArena Hard Prompts	1417	1379
ForecastBench	58	59.2
SimpleBench	40%	—
DTBench	82.7%	—
LMCA	24.3%	—
Epoch Capabilities Index	139.92	—

Math DeepSeek-V3.1 leads

DeepSeek-V3.1: 38.9 (#122), GLM-4.5-Air: 36.2 (#170)

Math benchmarks
Benchmark	DeepSeek-V3.1	GLM-4.5-Air
LMArena Math	1420	1396
Omni-MATH	—	39.1%

Knowledge DeepSeek-V3.1 leads

DeepSeek-V3.1: 43.7 (#90), GLM-4.5-Air: 35.0 (#191)

Knowledge benchmarks
Benchmark	DeepSeek-V3.1	GLM-4.5-Air
Vectara Hallucination Rate	5.5%	9.3%
LMArena Expert	1405	1370
Humanity's Last Exam	—	8.1%
MMLU-Pro	—	76.2%
GPQA (HELM)	—	59.4%

Multilingual DeepSeek-V3.1 leads

DeepSeek-V3.1: 51.6 (#106), GLM-4.5-Air: 49.1 (#135)

Multilingual benchmarks
Benchmark	DeepSeek-V3.1	GLM-4.5-Air
LMArena Non-English	1400	1366
LMArena Chinese	1469	1426
LMArena French	1447	1399
LMArena German	1411	1377
LMArena Japanese	1378	1348
LMArena Korean	1337	1308
LMArena Russian	1405	1373
LMArena Spanish	1431	1386

Instruction Following DeepSeek-V3.1 leads

DeepSeek-V3.1: 73.9 (#110), GLM-4.5-Air: 69.6 (#171)

Instruction Following benchmarks
Benchmark	DeepSeek-V3.1	GLM-4.5-Air
LMArena Instruction Following	1400	1354
IFEval	—	81.2%

Long Context GLM-4.5-Air leads

DeepSeek-V3.1: 36.3 (#232), GLM-4.5-Air: 41.6 (#135)

Long Context benchmarks
Benchmark	DeepSeek-V3.1	GLM-4.5-Air
LMArena Longer Query	1422	1366
Fiction.LiveBench	52.8%	—

Writing & Preference DeepSeek-V3.1 leads

DeepSeek-V3.1: 60.3 (#98), GLM-4.5-Air: 55.9 (#139)

Writing & Preference benchmarks
Benchmark	DeepSeek-V3.1	GLM-4.5-Air
LMArena Text	1420	1384
LMArena Creative Writing	1401	1343
LMArena Multi-Turn	1408	1371
EQ-Bench Creative Writing	1436	—
WildBench	—	78.9%

Frequently asked questions

Is DeepSeek-V3.1 better than GLM-4.5-Air?

DeepSeek-V3.1 is the stronger model overall, scoring 42.8 to 38.9 on the Noometry Index.

Which is cheaper, DeepSeek-V3.1 or GLM-4.5-Air?

DeepSeek-V3.1 is cheaper. It lists at $0.25 per million input tokens and $0.95 per million output tokens; GLM-4.5-Air lists at $0.20 and $1.10.

Is DeepSeek-V3.1 or GLM-4.5-Air better for coding?

DeepSeek-V3.1 scores higher on coding benchmarks: 40.3 versus 33.3 in the Noometry coding category.

Which has the bigger context window?

DeepSeek-V3.1 does, with 164K tokens against 131K.

How many benchmarks do DeepSeek-V3.1 and GLM-4.5-Air share?

20 benchmarks have published results for both models. DeepSeek-V3.1 has 27 scored results on Noometry and GLM-4.5-Air has 27.