LLMBenchmark.ai

Enter opens the highlighted result. Esc closes.

DeepSeek V3.2 vs GLM 5.2

Updated October 21, 2026. 62 benchmarks have a result for both. Coding is ranked on the Vibe Coding Index; every other board is shown as published.

Replace DeepSeek V3.2 Replace GLM 5.2
DeepSeek V3.2DeepSeek
Coding
54.4
Overall
21.5
Output $/M
$0.42
1
GLM 5.2Z.ai
Coding
57.7
Overall
33.7
Output $/M
$7.00
61
vs boards won of 62 shared

Decision reading

Pick GLM 5.2 for coding: it scores 57.7 on the Vibe Coding Index against 54.4 for DeepSeek V3.2, though on 1 benchmark against 2.

Pick DeepSeek V3.2 when output tokens cost 16.7x less, or when the benchmark that matters is LiveCodeBench (Vals AI), where it leads.

Model ADeepSeek V3.2DeepSeek54.4/1002 independent benchmarks$0.99 per solved taskModel BGLM 5.2Z.ai57.7/1001 independent benchmark$0.86 per solved task
Shared results
62
DeepSeek V3.2 only
13
GLM 5.2 only
35
Like-for-like categories
8 of 9

Shared results can support a head-to-head reading. Results only one model has describe coverage, not superiority.

Which one for your work

Each card is decided by one number both models carry. Where only one has it, the card says so and makes no call.

At a glance

Coding index
DeepSeek V3.2 54.4 from 2 benchmarks, GLM 5.2 57.7 from 1 benchmark.
Widest category
Agentic work, 26.0 points apart on average over 3 shared benchmarks.
Widest single gap
Vibe Code Bench (Vals AI): DeepSeek V3.2 5.1% against GLM 5.2 64.0%, 58.9 points apart on a 0 to 100 scale.
Output price
DeepSeek V3.2 $0.42 against GLM 5.2 $7.00 per million output tokens, 16.7x apart.
Context window
DeepSeek V3.2 takes 164K tokens, GLM 5.2 1.05M.

Benchmark by benchmark

First the score: boards each model leads, per category, where both have a result. Then every board, a category to a fold; bars are each result on a 0 to 100 scale, so compare the two in a row, not one row with another.

Boards won, where both have a result
CategoryDeepSeek V3.2BarsGLM 5.2Ties
Coding19
Intelligence02
Agentic work03
Reasoning08
Math02
Knowledge08
Web design08
General021
All shared boards161
CodingLike-for-likeFour of these boards feed the Vibe Coding Index19
Coding benchmarks for DeepSeek V3.2 and GLM 5.2
BenchmarkDeepSeek V3.2GLM 5.2Leader
SWE-bench Verified70.0%Not measuredOnly DeepSeek V3.2 measured
SWE-bench Multilingual59.0%Not measuredOnly DeepSeek V3.2 measured
LMArena WebDev13621604GLM 5.2
AA Coding Index44.268.8GLM 5.2
LiveCodeBench (Vals AI)80.7%69.5%DeepSeek V3.2
SWE-bench Verified (Vals AI)67.6%82.8%GLM 5.2
Vibe Code Bench (Vals AI)5.1%64.0%GLM 5.2
Code migration (Vals AI)Not measured37.9%Only GLM 5.2 measured
ProgramBench (Vals AI)Not measured0.5%Only GLM 5.2 measured
Terminal-Bench 2.1 (Vals AI)Not measured67.8%Only GLM 5.2 measured
SkillsBench (Vals AI)Not measured45.1%Only GLM 5.2 measured
LiveCodeBench (Artificial Analysis run)86.2%Not measuredOnly DeepSeek V3.2 measured
SciCodeNot measured51.2%Only GLM 5.2 measured
Terminal-Bench Hard (Artificial Analysis run)35.6%50.8%GLM 5.2
Terminal-Bench39.6%Not measuredOnly DeepSeek V3.2 measured
FrontierCodeNot measured24.5%Only GLM 5.2 measured
WeirdMLNot measured70.1%Only GLM 5.2 measured
ALE-BenchNot measured1047.0Only GLM 5.2 measured
LMArena Text: coding14531487GLM 5.2
LMArena Text: software and IT14491486GLM 5.2
LMArena WebDev: React13501617GLM 5.2
LMArena WebDev: HTML13711565GLM 5.2
Neither model has a result on the other 16 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index02
Intelligence benchmarks for DeepSeek V3.2 and GLM 5.2
BenchmarkDeepSeek V3.2GLM 5.2Leader
AA Intelligence Index21.533.7GLM 5.2
Epoch Capabilities Index146.1151.9GLM 5.2
Agentic workLike-for-likePublic boards, shown as published, kept out of the index03
Agentic work benchmarks for DeepSeek V3.2 and GLM 5.2
BenchmarkDeepSeek V3.2GLM 5.2Leader
AA Agentic IndexNot measured38.4Only GLM 5.2 measured
Design Arena: agents building web appsNot measured1222Only GLM 5.2 measured
Design Arena: agents building full-stack appsNot measured1241Only GLM 5.2 measured
Design Arena: agents building mobile appsNot measured1160Only GLM 5.2 measured
Design Arena: agents building Android appsNot measured1184Only GLM 5.2 measured
Design Arena: agents building Godot gamesNot measured1142Only GLM 5.2 measured
Design Arena: agents building gamesNot measured1206Only GLM 5.2 measured
Design Arena: HTML slidesNot measured1175Only GLM 5.2 measured
Design Arena: PowerPoint slidesNot measured1195Only GLM 5.2 measured
SREBench (Vals AI)Not measured0.0%Only GLM 5.2 measured
Finance Agent (Vals AI)Not measured49.7%Only GLM 5.2 measured
Legal Research Bench (Vals AI)Not measured31.3%Only GLM 5.2 measured
Excel Modeling Benchmark (Vals AI)Not measured61.5%Only GLM 5.2 measured
τ²-Bench Telecom90.6%99.1%GLM 5.2
APEX-Agents21.3%45.2%GLM 5.2
PostTrainBenchNot measured31.7%Only GLM 5.2 measured
Vending-Bench 2$1,034$8,314GLM 5.2
Neither model has a result on the other 22 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index08
Reasoning benchmarks for DeepSeek V3.2 and GLM 5.2
BenchmarkDeepSeek V3.2GLM 5.2Leader
GPQA Diamond80.3%85.6%GLM 5.2
GPQA Diamond (Artificial Analysis run)84.0%89.5%GLM 5.2
Humanity's Last Exam (Artificial Analysis run)24.6%41.1%GLM 5.2
GPQA Diamond (Epoch AI run)Not measured91.9%Only GLM 5.2 measured
ARC-AGI57.0%77.0%GLM 5.2
ARC-AGI-24.0%22.8%GLM 5.2
SimpleBenchNot measured58.8%Only GLM 5.2 measured
Chess puzzlesNot measured22.0%Only GLM 5.2 measured
Mystery game puzzlesNot measured19.0%Only GLM 5.2 measured
EBR-benchNot measured9.5%Only GLM 5.2 measured
LMArena Text: hard prompts14341481GLM 5.2
LMArena Text: expert14361487GLM 5.2
LMArena Text: hard prompts in English14461481GLM 5.2
Neither model has a result on the other 8 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index02
Math benchmarks for DeepSeek V3.2 and GLM 5.2
BenchmarkDeepSeek V3.2GLM 5.2Leader
AIME84.6%Not measuredOnly DeepSeek V3.2 measured
MGSM (Vals AI)86.0%Not measuredOnly DeepSeek V3.2 measured
AA Math Index92.0Not measuredOnly DeepSeek V3.2 measured
AIME 2025 (Artificial Analysis run)92.0%Not measuredOnly DeepSeek V3.2 measured
OTIS Mock AIMENot measured86.4%Only GLM 5.2 measured
FrontierMath, tiers 1 to 3Not measured59.2%Only GLM 5.2 measured
FrontierMath, tier 4Not measured29.3%Only GLM 5.2 measured
ProofBenchNot measured35.0%Only GLM 5.2 measured
LMArena Text: math14351483GLM 5.2
LMArena Text: mathematical work14401489GLM 5.2
Neither model has a result on the other 3 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index08
Knowledge benchmarks for DeepSeek V3.2 and GLM 5.2
BenchmarkDeepSeek V3.2GLM 5.2Leader
MMLU Pro84.9%86.7%GLM 5.2
LegalBench (Vals AI)76.1%84.1%GLM 5.2
CaseLaw (Vals AI)55.4%Not measuredOnly DeepSeek V3.2 measured
CorpFin (Vals AI)51.0%66.1%GLM 5.2
TaxEval (Vals AI)68.2%73.3%GLM 5.2
MedQA (Vals AI)93.9%Not measuredOnly DeepSeek V3.2 measured
MedCode (Vals AI)Not measured40.8%Only GLM 5.2 measured
MMLU Pro (Artificial Analysis run)86.2%Not measuredOnly DeepSeek V3.2 measured
SimpleQA VerifiedNot measured34.2%Only GLM 5.2 measured
LMArena Text: science14461489GLM 5.2
LMArena Text: medicine and healthcare14401475GLM 5.2
LMArena Text: legal and government14251486GLM 5.2
LMArena Text: business and finance14101451GLM 5.2
Neither model has a result on the other 4 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index08
Web design benchmarks for DeepSeek V3.2 and GLM 5.2
BenchmarkDeepSeek V3.2GLM 5.2Leader
Design Arena: websites11751285GLM 5.2
Design Arena: UI components11491296GLM 5.2
Design Arena: game dev11321268GLM 5.2
Design Arena: data viz11581285GLM 5.2
Design Arena: 3D11331288GLM 5.2
Design Arena: all code categories11641288GLM 5.2
Design Arena: SVG10321214GLM 5.2
Design Arena: ASCII art10821190GLM 5.2
Neither model has a result on the other 1 board here
GeneralLike-for-likePublic boards, shown as published, kept out of the index021
General benchmarks for DeepSeek V3.2 and GLM 5.2
BenchmarkDeepSeek V3.2GLM 5.2Leader
MedScribe (Vals AI)Not measured83.5%Only GLM 5.2 measured
IFBench60.7%73.3%GLM 5.2
Long context reasoning (AA-LCR)73.3%78.3%GLM 5.2
DTBench85.6%93.6%GLM 5.2
LMCA28.8%45.8%GLM 5.2
CL-bench12.4%Not measuredOnly DeepSeek V3.2 measured
Surface Evolver BenchNot measured55.6%Only GLM 5.2 measured
CL-bench Life7.4%Not measuredOnly DeepSeek V3.2 measured
LMArena Text14251470GLM 5.2
LMArena Text: instruction following14131466GLM 5.2
LMArena Text: creative writing14001463GLM 5.2
LMArena Text: English14391475GLM 5.2
LMArena Text: non-English14091459GLM 5.2
LMArena Text: Chinese14551522GLM 5.2
LMArena Text: German14131470GLM 5.2
LMArena Text: Japanese13741450GLM 5.2
LMArena Text: Korean13701440GLM 5.2
LMArena Text: French14291479GLM 5.2
LMArena Text: Spanish14241478GLM 5.2
LMArena Text: Russian14241468GLM 5.2
LMArena Text: Polish14191456GLM 5.2
LMArena Text: multi-turn14271469GLM 5.2
LMArena Text: longer queries14281480GLM 5.2
LMArena Text: writing and language14071463GLM 5.2
LMArena Text: entertainment and media13951450GLM 5.2
Neither model has a result on the other 21 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for DeepSeek V3.2 and GLM 5.2
BenchmarkDeepSeek V3.2GLM 5.2Leader
Neither model has a result on 5 boards here

What each workload costs

Three fixed token mixes at each model's listed price, so per-token rates become a bill.

Chat turn

1K input + 500 output tokens

DeepSeek V3.2$0.0005Fits in one request
GLM 5.2$0.0035Fits in one request

DeepSeek V3.2 costs less

Repository review

50K input + 3K output tokens

DeepSeek V3.2$0.015Fits in one request
GLM 5.2$0.023Fits in one request

DeepSeek V3.2 costs less

Agent loop

200K input + 10K output tokens

DeepSeek V3.2$0.060Exceeds the context window
GLM 5.2$0.077Fits in one request

DeepSeek V3.2 costs less

Listed OpenRouter rates, without cache or batch discounts. A workload that does not fit a model's context window would need splitting.

Specification differences

The leading value in a row is marked; a row with a missing value names no leader. Our two rows are marked in red.

Specifications of DeepSeek V3.2 and GLM 5.2
SpecificationDeepSeek V3.2Z.ai GLM 5.2
MakerDeepSeekZ.ai
WeightsOpen weightsOpen weights
ReleasedDecember 1, 2025June 16, 2026
Context window164K1.05M
Input $/Mtok$0.28$0.0341
Output $/Mtok$0.42$7.00
Coding index54.457.7
Independent measurements behind it21
Cost per solved task$0.99$0.86
SourceOpenRouter listingOpenRouter listing

Questions

Is DeepSeek V3.2 better than GLM 5.2 for coding?

Pick GLM 5.2 for coding: it scores 57.7 on the Vibe Coding Index against 54.4 for DeepSeek V3.2, though on 1 benchmark against 2. Pick DeepSeek V3.2 when output tokens cost 16.7x less, or when the benchmark that matters is LiveCodeBench (Vals AI), where it leads.

Which is cheaper, DeepSeek V3.2 or GLM 5.2?

DeepSeek V3.2 costs $0.28 per million input tokens and $0.42 per million output tokens; GLM 5.2 costs $0.0341 and $7.00. Per solved SWE-bench Verified task, DeepSeek V3.2 costs $0.99 and GLM 5.2 $0.86.

Which has the bigger context window, DeepSeek V3.2 or GLM 5.2?

GLM 5.2, with 1.05M tokens against 164K.

How many benchmarks is the DeepSeek V3.2 vs GLM 5.2 comparison based on?

62 benchmarks have a result for both models as of October 21, 2026. DeepSeek V3.2's coding index rests on 2 independent benchmarks and GLM 5.2's on 1. Where only one model has a result, the page says so instead of naming a winner.

Compare next