LLMBenchmark.ai

Enter opens the highlighted result. Esc closes.

DeepSeek V3.2 vs GLM 4.5

Updated October 21, 2026. 56 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 4.5
DeepSeek V3.2DeepSeek
Coding
54.4
Overall
21.5
Output $/M
$0.42
36
GLM 4.5Z.ai
Coding
64.2
Overall
12.8
Output $/M
$2.20
20
vs boards won of 56 shared

Decision reading

Pick GLM 4.5 for coding: it scores 64.2 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 5.2x less, or when the context window is larger, 164K against 131K tokens.

Model ADeepSeek V3.2DeepSeek54.4/1002 independent benchmarks$0.99 per solved taskModel BGLM 4.5Z.ai64.2/1001 independent benchmarkNo cost per solved task
Shared results
56
DeepSeek V3.2 only
19
GLM 4.5 only
4
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 4.5 64.2 from 1 benchmark.
Widest category
Agentic work, 47.7 points apart on average over 1 shared benchmark.
Widest single gap
τ²-Bench Telecom: DeepSeek V3.2 90.6% against GLM 4.5 43.0%, 47.7 points apart on a 0 to 100 scale.
Output price
DeepSeek V3.2 $0.42 against GLM 4.5 $2.20 per million output tokens, 5.2x apart.
Context window
DeepSeek V3.2 takes 164K tokens, GLM 4.5 131K.

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 4.5Ties
Coding60
Intelligence10
Agentic work10
Reasoning60
Math42
Knowledge55
Web design25
General118
All shared boards3620
CodingLike-for-likeFour of these boards feed the Vibe Coding Index60
Coding benchmarks for DeepSeek V3.2 and GLM 4.5
BenchmarkDeepSeek V3.2GLM 4.5Leader
SWE-bench Verified70.0%64.2%DeepSeek V3.2
SWE-bench Multilingual59.0%Not measuredOnly DeepSeek V3.2 measured
LMArena WebDev1362Not measuredOnly DeepSeek V3.2 measured
AA Coding Index44.2Not measuredOnly DeepSeek V3.2 measured
LiveCodeBench (Vals AI)80.7%67.5%DeepSeek V3.2
SWE-bench Verified (Vals AI)67.6%Not measuredOnly DeepSeek V3.2 measured
Vibe Code Bench (Vals AI)5.1%Not measuredOnly DeepSeek V3.2 measured
LiveCodeBench (Artificial Analysis run)86.2%73.8%DeepSeek V3.2
Terminal-Bench Hard (Artificial Analysis run)35.6%22.0%DeepSeek V3.2
Terminal-Bench39.6%Not measuredOnly DeepSeek V3.2 measured
WeirdMLNot measured40.6%Only GLM 4.5 measured
ALE-BenchNot measured344.8Only GLM 4.5 measured
LMArena Text: coding14531434DeepSeek V3.2
LMArena Text: software and IT14491440DeepSeek V3.2
LMArena WebDev: React1350Not measuredOnly DeepSeek V3.2 measured
LMArena WebDev: HTML1371Not measuredOnly DeepSeek V3.2 measured
Neither model has a result on the other 22 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index10
Intelligence benchmarks for DeepSeek V3.2 and GLM 4.5
BenchmarkDeepSeek V3.2GLM 4.5Leader
AA Intelligence Index21.512.8DeepSeek V3.2
Epoch Capabilities Index146.1Not measuredOnly DeepSeek V3.2 measured
Agentic workLike-for-likePublic boards, shown as published, kept out of the index10
Agentic work benchmarks for DeepSeek V3.2 and GLM 4.5
BenchmarkDeepSeek V3.2GLM 4.5Leader
τ²-Bench Telecom90.6%43.0%DeepSeek V3.2
APEX-Agents21.3%Not measuredOnly DeepSeek V3.2 measured
Vending-Bench 2$1,034Not measuredOnly DeepSeek V3.2 measured
Neither model has a result on the other 36 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index60
Reasoning benchmarks for DeepSeek V3.2 and GLM 4.5
BenchmarkDeepSeek V3.2GLM 4.5Leader
GPQA Diamond80.3%72.2%DeepSeek V3.2
GPQA Diamond (Artificial Analysis run)84.0%78.2%DeepSeek V3.2
Humanity's Last Exam (Artificial Analysis run)24.6%13.0%DeepSeek V3.2
ARC-AGI57.0%Not measuredOnly DeepSeek V3.2 measured
ARC-AGI-24.0%Not measuredOnly DeepSeek V3.2 measured
LMArena Text: hard prompts14341429DeepSeek V3.2
LMArena Text: expert14361433DeepSeek V3.2
LMArena Text: hard prompts in English14461429DeepSeek V3.2
Neither model has a result on the other 13 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index42
Math benchmarks for DeepSeek V3.2 and GLM 4.5
BenchmarkDeepSeek V3.2GLM 4.5Leader
AIME84.6%86.7%GLM 4.5
MATH 500 (Vals AI)Not measured94.0%Only GLM 4.5 measured
MGSM (Vals AI)86.0%90.8%GLM 4.5
AA Math Index92.073.7DeepSeek V3.2
AIME 2025 (Artificial Analysis run)92.0%73.7%DeepSeek V3.2
LMArena Text: math14351427DeepSeek V3.2
LMArena Text: mathematical work14401430DeepSeek V3.2
Neither model has a result on the other 6 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index55
Knowledge benchmarks for DeepSeek V3.2 and GLM 4.5
BenchmarkDeepSeek V3.2GLM 4.5Leader
MMLU Pro84.9%81.2%DeepSeek V3.2
LegalBench (Vals AI)76.1%75.6%DeepSeek V3.2
CaseLaw (Vals AI)55.4%Not measuredOnly DeepSeek V3.2 measured
CorpFin (Vals AI)51.0%61.0%GLM 4.5
TaxEval (Vals AI)68.2%72.4%GLM 4.5
MedQA (Vals AI)93.9%90.0%DeepSeek V3.2
MMLU Pro (Artificial Analysis run)86.2%83.5%DeepSeek V3.2
LMArena Text: science14461444DeepSeek V3.2
LMArena Text: medicine and healthcare14401455GLM 4.5
LMArena Text: legal and government14251438GLM 4.5
LMArena Text: business and finance14101420GLM 4.5
Neither model has a result on the other 6 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index25
Web design benchmarks for DeepSeek V3.2 and GLM 4.5
BenchmarkDeepSeek V3.2GLM 4.5Leader
Design Arena: websites11751170DeepSeek V3.2
Design Arena: UI components11491144DeepSeek V3.2
Design Arena: game dev11321145GLM 4.5
Design Arena: data viz11581159GLM 4.5
Design Arena: 3D11331170GLM 4.5
Design Arena: all code categories11641165GLM 4.5
Design Arena: SVG10321096GLM 4.5
Design Arena: ASCII art1082Not measuredOnly DeepSeek V3.2 measured
Neither model has a result on the other 1 board here
GeneralLike-for-likePublic boards, shown as published, kept out of the index118
General benchmarks for DeepSeek V3.2 and GLM 4.5
BenchmarkDeepSeek V3.2GLM 4.5Leader
IFBench60.7%44.1%DeepSeek V3.2
Long context reasoning (AA-LCR)73.3%52.7%DeepSeek V3.2
DTBench85.6%Not measuredOnly DeepSeek V3.2 measured
LMCA28.8%Not measuredOnly DeepSeek V3.2 measured
Creative writing (Lech Mazur)Not measured73.4%Only GLM 4.5 measured
CL-bench12.4%Not measuredOnly DeepSeek V3.2 measured
CL-bench Life7.4%Not measuredOnly DeepSeek V3.2 measured
LMArena Text14251430GLM 4.5
LMArena Text: instruction following14131404DeepSeek V3.2
LMArena Text: creative writing14001395DeepSeek V3.2
LMArena Text: English14391437DeepSeek V3.2
LMArena Text: non-English14091417GLM 4.5
LMArena Text: Chinese14551465GLM 4.5
LMArena Text: German14131407DeepSeek V3.2
LMArena Text: Japanese13741415GLM 4.5
LMArena Text: Korean13701380GLM 4.5
LMArena Text: French14291418DeepSeek V3.2
LMArena Text: Spanish14241454GLM 4.5
LMArena Text: Russian14241413DeepSeek V3.2
LMArena Text: Polish14191422GLM 4.5
LMArena Text: multi-turn14271415DeepSeek V3.2
LMArena Text: longer queries14281412DeepSeek V3.2
LMArena Text: writing and language14071403DeepSeek V3.2
LMArena Text: entertainment and media13951404GLM 4.5
Neither model has a result on the other 22 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 4.5
BenchmarkDeepSeek V3.2GLM 4.5Leader
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 4.5$0.0017Fits in one request

DeepSeek V3.2 costs less

Repository review

50K input + 3K output tokens

DeepSeek V3.2$0.015Fits in one request
GLM 4.5$0.037Fits in one request

DeepSeek V3.2 costs less

Agent loop

200K input + 10K output tokens

DeepSeek V3.2$0.060Exceeds the context window
GLM 4.5$0.142Exceeds the context window

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 4.5
SpecificationDeepSeek V3.2Z.ai GLM 4.5
MakerDeepSeekZ.ai
WeightsOpen weightsOpen weights
ReleasedDecember 1, 2025July 25, 2025
Context window164K131K
Input $/Mtok$0.28$0.60
Output $/Mtok$0.42$2.20
Coding index54.464.2
Independent measurements behind it21
Cost per solved task$0.99Not computable
SourceOpenRouter listingOpenRouter listing

Questions

Is DeepSeek V3.2 better than GLM 4.5 for coding?

Pick GLM 4.5 for coding: it scores 64.2 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 5.2x less, or when the context window is larger, 164K against 131K tokens.

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

DeepSeek V3.2 costs $0.28 per million input tokens and $0.42 per million output tokens; GLM 4.5 costs $0.60 and $2.20. Cost per solved task is known for DeepSeek V3.2 only.

Which has the bigger context window, DeepSeek V3.2 or GLM 4.5?

DeepSeek V3.2, with 164K tokens against 131K.

How many benchmarks is the DeepSeek V3.2 vs GLM 4.5 comparison based on?

56 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 4.5's on 1. Where only one model has a result, the page says so instead of naming a winner.

Compare next