Enter opens the highlighted result. Esc closes.

DeepSeek V3.1 Terminus vs GLM 4.5

DeepSeek V3.1 Terminus leads GLM 4.5 on the Artificial Analysis Intelligence Index, 14.8 to 12.8, and DeepSeek V3.1 Terminus costs 2.2x less per output token ($1.0142 vs $2.20 per million). Updated October 21, 2026; 40 benchmarks have a result for both.

Replace DeepSeek V3.1 Terminus Replace GLM 4.5
DeepSeek V3.1 TerminusDeepSeek
Coding
No data
Overall
No data
Output $/M
$1.0142
24
GLM 4.5Z.ai
Coding
No data
Overall
No data
Output $/M
$2.20
16
vs boards won of 40 shared

DeepSeek V3.1 Terminus or GLM 4.5: which to pick

Neither model has a coding score yet, so the two cannot be ranked on coding.

Pick DeepSeek V3.1 Terminus for output tokens 2.2x cheaper ($1.0142 vs $2.20 per million) and a larger context window (164K against 131K tokens).

Model ADeepSeek V3.1 TerminusDeepSeekNot benchmarkedNo coding benchmarkNo cost per solved taskModel BGLM 4.5Z.aiNot benchmarkedNo coding benchmarkNo cost per solved task
Shared results
40
DeepSeek V3.1 Terminus only
5
GLM 4.5 only
21
Like-for-like categories
8 of 9

Only benchmarks that tested both models count toward a pick; a result only one model has is shown but not compared.

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 V3.1 Terminus not benchmarked, GLM 4.5 not benchmarked.
Widest category
Math, 12.1 points apart on average over 3 shared benchmarks.
Widest single gap
Long context reasoning (AA-LCR): DeepSeek V3.1 Terminus 69.3% against GLM 4.5 52.7%, 16.7 points apart on a 0 to 100 scale.
Output price
DeepSeek V3.1 Terminus $1.0142 against GLM 4.5 $2.20 per million output tokens, 2.2x apart.
Context window
DeepSeek V3.1 Terminus 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 V3.1 TerminusBarsGLM 4.5Ties
Coding32
Intelligence10
Agentic work01
Reasoning32
Math21
Knowledge41
Web design43
General76
All shared boards2416
CodingLike-for-likeSix of these boards feed the Vibe Coding Index32
Coding benchmarks for DeepSeek V3.1 Terminus and GLM 4.5
BenchmarkDeepSeek V3.1 TerminusGLM 4.5Leader
SWE-bench VerifiedNot measured64.2%Only GLM 4.5 measured
AA Coding Index43.5Not measuredOnly DeepSeek V3.1 Terminus measured
LiveCodeBench (Vals AI)Not measured67.5%Only GLM 4.5 measured
LiveCodeBench (Artificial Analysis run)79.8%73.8%DeepSeek V3.1 Terminus
SciCode38.0%Not measuredOnly DeepSeek V3.1 Terminus measured
Terminal-Bench Hard (Artificial Analysis run)31.8%22.0%DeepSeek V3.1 Terminus
WeirdMLNot measured40.6%Only GLM 4.5 measured
ALE-Bench745.2344.8DeepSeek V3.1 Terminus
LMArena Text: coding14251434GLM 4.5
LMArena Text: software and IT14261440GLM 4.5
Neither model has a result on the other 29 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index10
Intelligence benchmarks for DeepSeek V3.1 Terminus and GLM 4.5
BenchmarkDeepSeek V3.1 TerminusGLM 4.5Leader
AA Intelligence Index14.812.8DeepSeek V3.1 Terminus
Neither model has a result on the other 1 board here
Agentic workLike-for-likePublic boards, shown as published, kept out of the index01
Agentic work benchmarks for DeepSeek V3.1 Terminus and GLM 4.5
BenchmarkDeepSeek V3.1 TerminusGLM 4.5Leader
AA Agentic Index6.8Not measuredOnly DeepSeek V3.1 Terminus measured
τ²-Bench Telecom37.1%43.0%GLM 4.5
Neither model has a result on the other 39 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index32
Reasoning benchmarks for DeepSeek V3.1 Terminus and GLM 4.5
BenchmarkDeepSeek V3.1 TerminusGLM 4.5Leader
GPQA DiamondNot measured72.2%Only GLM 4.5 measured
GPQA Diamond (Artificial Analysis run)79.2%78.2%DeepSeek V3.1 Terminus
Humanity's Last Exam (Artificial Analysis run)16.4%13.0%DeepSeek V3.1 Terminus
LMArena Text: hard prompts14261429GLM 4.5
LMArena Text: expertNot measured1433Only GLM 4.5 measured
LMArena Text: hard prompts in English14321429DeepSeek V3.1 Terminus
Kagi LLM Benchmark57.4%57.9%GLM 4.5
Neither model has a result on the other 15 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index21
Math benchmarks for DeepSeek V3.1 Terminus and GLM 4.5
BenchmarkDeepSeek V3.1 TerminusGLM 4.5Leader
AIMENot measured86.7%Only GLM 4.5 measured
MATH 500 (Vals AI)Not measured94.0%Only GLM 4.5 measured
MGSM (Vals AI)Not measured90.8%Only GLM 4.5 measured
AA Math Index89.773.7DeepSeek V3.1 Terminus
AIME 2025 (Artificial Analysis run)89.7%73.7%DeepSeek V3.1 Terminus
LMArena Text: math14021427GLM 4.5
LMArena Text: mathematical workNot measured1430Only GLM 4.5 measured
Neither model has a result on the other 6 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index41
Knowledge benchmarks for DeepSeek V3.1 Terminus and GLM 4.5
BenchmarkDeepSeek V3.1 TerminusGLM 4.5Leader
MMLU ProNot measured81.2%Only GLM 4.5 measured
LegalBench (Vals AI)Not measured75.6%Only GLM 4.5 measured
CorpFin (Vals AI)Not measured61.0%Only GLM 4.5 measured
TaxEval (Vals AI)Not measured72.4%Only GLM 4.5 measured
MedQA (Vals AI)Not measured90.0%Only GLM 4.5 measured
MMLU Pro (Artificial Analysis run)85.1%83.5%DeepSeek V3.1 Terminus
LMArena Text: science14511444DeepSeek V3.1 Terminus
LMArena Text: medicine and healthcare14681455DeepSeek V3.1 Terminus
LMArena Text: legal and government14571438DeepSeek V3.1 Terminus
LMArena Text: business and finance14141420GLM 4.5
Neither model has a result on the other 8 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index43
Web design benchmarks for DeepSeek V3.1 Terminus and GLM 4.5
BenchmarkDeepSeek V3.1 TerminusGLM 4.5Leader
Design Arena: websites11871170DeepSeek V3.1 Terminus
Design Arena: UI components11821144DeepSeek V3.1 Terminus
Design Arena: game dev11311145GLM 4.5
Design Arena: data viz11671159DeepSeek V3.1 Terminus
Design Arena: 3D11411170GLM 4.5
Design Arena: all code categories11751165DeepSeek V3.1 Terminus
Design Arena: SVG10651096GLM 4.5
Neither model has a result on the other 2 boards here
GeneralLike-for-likePublic boards, shown as published, kept out of the index76
General benchmarks for DeepSeek V3.1 Terminus and GLM 4.5
BenchmarkDeepSeek V3.1 TerminusGLM 4.5Leader
IFBench57.0%44.1%DeepSeek V3.1 Terminus
Long context reasoning (AA-LCR)69.3%52.7%DeepSeek V3.1 Terminus
DTBench81.3%Not measuredOnly DeepSeek V3.1 Terminus measured
LMCA28.6%Not measuredOnly DeepSeek V3.1 Terminus measured
Creative writing (Lech Mazur)Not measured73.4%Only GLM 4.5 measured
LMArena Text14191430GLM 4.5
LMArena Text: instruction following14041404GLM 4.5
LMArena Text: creative writing14031395DeepSeek V3.1 Terminus
LMArena Text: English14291437GLM 4.5
LMArena Text: non-English14081417GLM 4.5
LMArena Text: ChineseNot measured1465Only GLM 4.5 measured
LMArena Text: GermanNot measured1407Only GLM 4.5 measured
LMArena Text: JapaneseNot measured1415Only GLM 4.5 measured
LMArena Text: KoreanNot measured1380Only GLM 4.5 measured
LMArena Text: FrenchNot measured1418Only GLM 4.5 measured
LMArena Text: SpanishNot measured1454Only GLM 4.5 measured
LMArena Text: Russian14351413DeepSeek V3.1 Terminus
LMArena Text: Polish14271422DeepSeek V3.1 Terminus
LMArena Text: multi-turn14111415GLM 4.5
LMArena Text: longer queries14221412DeepSeek V3.1 Terminus
LMArena Text: writing and language14091403DeepSeek V3.1 Terminus
LMArena Text: entertainment and media13911404GLM 4.5
Neither model has a result on the other 24 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for DeepSeek V3.1 Terminus and GLM 4.5
BenchmarkDeepSeek V3.1 TerminusGLM 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 V3.1 Terminus$0.0008Fits in one request
GLM 4.5$0.0017Fits in one request

DeepSeek V3.1 Terminus costs less

Repository review

50K input + 3K output tokens

DeepSeek V3.1 Terminus$0.018Fits in one request
GLM 4.5$0.037Fits in one request

DeepSeek V3.1 Terminus costs less

Agent loop

200K input + 10K output tokens

DeepSeek V3.1 Terminus$0.071Exceeds the context window
GLM 4.5$0.142Exceeds the context window

DeepSeek V3.1 Terminus 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 V3.1 Terminus and GLM 4.5
SpecificationDeepSeek V3.1 TerminusZ.ai GLM 4.5
MakerDeepSeekZ.ai
WeightsOpen weightsOpen weights
ReleasedSeptember 22, 2025July 25, 2025
Context window164K131K
Input $/Mtok$0.3063$0.60
Output $/Mtok$1.0142$2.20
Coding indexNot benchmarkedNot benchmarked
Independent measurements behind it00
Cost per solved taskNot computableNot computable
SourceOpenRouter listingOpenRouter listing

Questions

Is DeepSeek V3.1 Terminus better than GLM 4.5 for coding?

Neither model has a coding score yet, so the two cannot be ranked on coding. Pick DeepSeek V3.1 Terminus for output tokens 2.2x cheaper ($1.0142 vs $2.20 per million) and a larger context window (164K against 131K tokens).

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

DeepSeek V3.1 Terminus costs $0.3063 per million input tokens and $1.0142 per million output tokens; GLM 4.5 costs $0.60 and $2.20. Cost per solved task is known for neither of them.

Which has the bigger context window, DeepSeek V3.1 Terminus or GLM 4.5?

DeepSeek V3.1 Terminus, with 164K tokens against 131K.

How many benchmarks is the DeepSeek V3.1 Terminus vs GLM 4.5 comparison based on?

40 benchmarks have a result for both models as of October 21, 2026. DeepSeek V3.1 Terminus's coding index rests on 0 independent benchmarks and GLM 4.5's on 0. Where only one model has a result, the page says so instead of naming a winner.

Compare next