LLMBenchmark.ai

Enter opens the highlighted result. Esc closes.

DeepSeek V4 Pro 0813 vs GLM 5.1

Updated October 21, 2026. 60 benchmarks have a result for both. Coding is ranked on the Vibe Coding Index; every other board is shown as published.

Replace DeepSeek V4 Pro 0813 Replace GLM 5.1
DeepSeek V4 Pro 0813DeepSeek
Coding
54.6
Overall
36.0
Output $/M
$1.98
33
GLM 5.1Z.ai
Coding
44.0
Overall
26.1
Output $/M
$4.40
26
vs boards won of 60 shared, 1 tied

Decision reading

Pick DeepSeek V4 Pro 0813 for coding: it scores 54.6 on the Vibe Coding Index against 44.0 for GLM 5.1.

Pick GLM 5.1 when what matters is Terminal-Bench 2.1 (Vals AI) or LMArena Text: coding, where it leads.

Model ADeepSeek V4 Pro 0813DeepSeek54.6/1001 independent benchmark$0.10 per solved taskModel BGLM 5.1Z.ai44.0/1001 independent benchmark$0.60 per solved task
Shared results
60
DeepSeek V4 Pro 0813 only
18
GLM 5.1 only
29
Like-for-like categories
7 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 V4 Pro 0813 54.6 from 1 benchmark, GLM 5.1 44.0 from 1 benchmark.
Widest category
Math, 14.1 points apart on average over 5 shared benchmarks.
Widest single gap
Vibe Code Bench (Vals AI): DeepSeek V4 Pro 0813 82.3% against GLM 5.1 31.5%, 50.8 points apart on a 0 to 100 scale.
Output price
DeepSeek V4 Pro 0813 $1.98 against GLM 5.1 $4.40 per million output tokens, 2.2x apart.
Context window
DeepSeek V4 Pro 0813 takes 1.05M tokens, GLM 5.1 205K.

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 V4 Pro 0813BarsGLM 5.1Ties
Coding1131 tie
Intelligence20
Agentic work40
Reasoning53
Math32
Knowledge55
General313
All shared boards33261 tie
CodingLike-for-likeFour of these boards feed the Vibe Coding Index113
Coding benchmarks for DeepSeek V4 Pro 0813 and GLM 5.1
BenchmarkDeepSeek V4 Pro 0813GLM 5.1Leader
LMArena WebDev15821508DeepSeek V4 Pro 0813
AA Coding Index68.855.8DeepSeek V4 Pro 0813
LiveCodeBench (Vals AI)87.5%81.4%DeepSeek V4 Pro 0813
SWE-bench Verified (Vals AI)96.4%76.4%DeepSeek V4 Pro 0813
Terminal-Bench 4 (Vals AI)14.1%Not measuredOnly DeepSeek V4 Pro 0813 measured
Vibe Code Bench (Vals AI)82.3%31.5%DeepSeek V4 Pro 0813
Code migration (Vals AI)41.5%25.8%DeepSeek V4 Pro 0813
ProgramBench (Vals AI)0.0%0.0%Tie
IOI (Vals AI)51.6%Not measuredOnly DeepSeek V4 Pro 0813 measured
Terminal-Bench 2.1 (Vals AI)54.7%56.9%GLM 5.1
Vibe Code Bench 1-100 (Vals AI)17.5%Not measuredOnly DeepSeek V4 Pro 0813 measured
SkillsBench (Vals AI)53.8%Not measuredOnly DeepSeek V4 Pro 0813 measured
CyberBench (Vals AI)72.9%Not measuredOnly DeepSeek V4 Pro 0813 measured
SciCode51.0%44.8%DeepSeek V4 Pro 0813
Terminal-Bench Hard (Artificial Analysis run)Not measured43.2%Only GLM 5.1 measured
FrontierCode28.5%Not measuredOnly DeepSeek V4 Pro 0813 measured
WeirdML66.2%57.1%DeepSeek V4 Pro 0813
ALE-Bench1403.2887.1DeepSeek V4 Pro 0813
LMArena Text: coding14691485GLM 5.1
LMArena Text: software and IT14621484GLM 5.1
LMArena WebDev: React15881505DeepSeek V4 Pro 0813
LMArena WebDev: HTML15661508DeepSeek V4 Pro 0813
Neither model has a result on the other 16 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index20
Intelligence benchmarks for DeepSeek V4 Pro 0813 and GLM 5.1
BenchmarkDeepSeek V4 Pro 0813GLM 5.1Leader
AA Intelligence Index36.026.1DeepSeek V4 Pro 0813
Epoch Capabilities Index155.6149.8DeepSeek V4 Pro 0813
Agentic workLike-for-likePublic boards, shown as published, kept out of the index40
Agentic work benchmarks for DeepSeek V4 Pro 0813 and GLM 5.1
BenchmarkDeepSeek V4 Pro 0813GLM 5.1Leader
AA Agentic Index41.323.9DeepSeek V4 Pro 0813
Design Arena: agents building web appsNot measured1175Only GLM 5.1 measured
Design Arena: agents building full-stack appsNot measured1167Only GLM 5.1 measured
Design Arena: agents building mobile appsNot measured1162Only GLM 5.1 measured
Design Arena: agents building Android appsNot measured1186Only GLM 5.1 measured
Design Arena: agents building Godot gamesNot measured1053Only GLM 5.1 measured
Design Arena: agents building gamesNot measured1164Only GLM 5.1 measured
Design Arena: HTML slidesNot measured1170Only GLM 5.1 measured
Design Arena: PowerPoint slidesNot measured1189Only GLM 5.1 measured
Design Arena: spreadsheetsNot measured1315Only GLM 5.1 measured
SREBench (Vals AI)1.9%Not measuredOnly DeepSeek V4 Pro 0813 measured
Finance Agent (Vals AI)50.4%44.8%DeepSeek V4 Pro 0813
Tax Agent Bench (Vals AI)26.8%Not measuredOnly DeepSeek V4 Pro 0813 measured
Legal Research Bench (Vals AI)40.9%27.9%DeepSeek V4 Pro 0813
Harvey's Legal Agent Benchmark (Vals AI)10.8%Not measuredOnly DeepSeek V4 Pro 0813 measured
Excel Modeling Benchmark (Vals AI)52.8%Not measuredOnly DeepSeek V4 Pro 0813 measured
Terminal-Bench-Science (Vals AI)4.3%Not measuredOnly DeepSeek V4 Pro 0813 measured
τ²-Bench TelecomNot measured97.7%Only GLM 5.1 measured
APEX-Agents47.3%40.9%DeepSeek V4 Pro 0813
Vending-Bench 2Not measured$5,634Only GLM 5.1 measured
Neither model has a result on the other 19 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index53
Reasoning benchmarks for DeepSeek V4 Pro 0813 and GLM 5.1
BenchmarkDeepSeek V4 Pro 0813GLM 5.1Leader
GPQA Diamond92.4%84.5%DeepSeek V4 Pro 0813
GPQA Diamond (Artificial Analysis run)92.8%86.8%DeepSeek V4 Pro 0813
Humanity's Last Exam (Artificial Analysis run)41.0%30.1%DeepSeek V4 Pro 0813
GPQA Diamond (Epoch AI run)91.7%89.9%DeepSeek V4 Pro 0813
ARC-AGI90.5%Not measuredOnly DeepSeek V4 Pro 0813 measured
ARC-AGI-261.3%Not measuredOnly DeepSeek V4 Pro 0813 measured
SimpleBenchNot measured55.1%Only GLM 5.1 measured
Chess puzzles46.0%20.0%DeepSeek V4 Pro 0813
Mystery game puzzles43.0%Not measuredOnly DeepSeek V4 Pro 0813 measured
LMArena Text: hard prompts14581473GLM 5.1
LMArena Text: expert14591478GLM 5.1
LMArena Text: hard prompts in English14631482GLM 5.1
Neither model has a result on the other 9 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index32
Math benchmarks for DeepSeek V4 Pro 0813 and GLM 5.1
BenchmarkDeepSeek V4 Pro 0813GLM 5.1Leader
AIMENot measured91.9%Only GLM 5.1 measured
OTIS Mock AIME98.6%93.3%DeepSeek V4 Pro 0813
FrontierMath, tiers 1 to 364.6%36.8%DeepSeek V4 Pro 0813
FrontierMath, tier 426.8%Not measuredOnly DeepSeek V4 Pro 0813 measured
ProofBench50.0%22.2%DeepSeek V4 Pro 0813
LMArena Text: math14411473GLM 5.1
LMArena Text: mathematical work14431470GLM 5.1
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 V4 Pro 0813 and GLM 5.1
BenchmarkDeepSeek V4 Pro 0813GLM 5.1Leader
MMLU Pro87.0%86.9%DeepSeek V4 Pro 0813
LegalBench (Vals AI)82.4%84.4%GLM 5.1
CaseLaw (Vals AI)Not measured51.5%Only GLM 5.1 measured
CorpFin (Vals AI)65.4%64.5%DeepSeek V4 Pro 0813
TaxEval (Vals AI)73.1%71.2%DeepSeek V4 Pro 0813
MedCode (Vals AI)42.5%41.6%DeepSeek V4 Pro 0813
SimpleQA Verified52.9%34.0%DeepSeek V4 Pro 0813
LMArena Text: science14571478GLM 5.1
LMArena Text: medicine and healthcare14461474GLM 5.1
LMArena Text: legal and government14591465GLM 5.1
LMArena Text: business and finance14421452GLM 5.1
Neither model has a result on the other 6 boards here
Web designDirectional onlyPublic boards, shown as published, kept out of the index8 boards, none shared
Web design benchmarks for DeepSeek V4 Pro 0813 and GLM 5.1
BenchmarkDeepSeek V4 Pro 0813GLM 5.1Leader
Design Arena: websitesNot measured1290Only GLM 5.1 measured
Design Arena: UI componentsNot measured1292Only GLM 5.1 measured
Design Arena: game devNot measured1265Only GLM 5.1 measured
Design Arena: data vizNot measured1366Only GLM 5.1 measured
Design Arena: 3DNot measured1336Only GLM 5.1 measured
Design Arena: all code categoriesNot measured1268Only GLM 5.1 measured
Design Arena: SVGNot measured1217Only GLM 5.1 measured
Design Arena: ASCII artNot measured1142Only GLM 5.1 measured
Neither model has a result on the other 1 board here
GeneralLike-for-likePublic boards, shown as published, kept out of the index313
General benchmarks for DeepSeek V4 Pro 0813 and GLM 5.1
BenchmarkDeepSeek V4 Pro 0813GLM 5.1Leader
Vals Index53.0%Not measuredOnly DeepSeek V4 Pro 0813 measured
Public Benefits Bench (Vals AI)Not measured61.8%Only GLM 5.1 measured
MedScribe (Vals AI)80.2%72.3%DeepSeek V4 Pro 0813
IFBenchNot measured76.3%Only GLM 5.1 measured
Long context reasoning (AA-LCR)80.3%73.7%DeepSeek V4 Pro 0813
DTBench93.9%Not measuredOnly DeepSeek V4 Pro 0813 measured
LMCA45.5%Not measuredOnly DeepSeek V4 Pro 0813 measured
ExploitBenchNot measured18.1%Only GLM 5.1 measured
LMArena Text14461463GLM 5.1
LMArena Text: instruction following14481453GLM 5.1
LMArena Text: creative writing14391453GLM 5.1
LMArena Text: English14571476GLM 5.1
LMArena Text: non-English14311448GLM 5.1
LMArena Text: Chinese14811515GLM 5.1
LMArena Text: GermanNot measured1467Only GLM 5.1 measured
LMArena Text: JapaneseNot measured1435Only GLM 5.1 measured
LMArena Text: Korean14491424DeepSeek V4 Pro 0813
LMArena Text: French14461470GLM 5.1
LMArena Text: Spanish14351468GLM 5.1
LMArena Text: Russian14481454GLM 5.1
LMArena Text: PolishNot measured1449Only GLM 5.1 measured
LMArena Text: multi-turn14541473GLM 5.1
LMArena Text: longer queries14581467GLM 5.1
LMArena Text: writing and language14431454GLM 5.1
LMArena Text: entertainment and media14241442GLM 5.1
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 V4 Pro 0813 and GLM 5.1
BenchmarkDeepSeek V4 Pro 0813GLM 5.1Leader
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 V4 Pro 0813$0.0016Fits in one request
GLM 5.1$0.0036Fits in one request

DeepSeek V4 Pro 0813 costs less

Repository review

50K input + 3K output tokens

DeepSeek V4 Pro 0813$0.039Fits in one request
GLM 5.1$0.083Fits in one request

DeepSeek V4 Pro 0813 costs less

Agent loop

200K input + 10K output tokens

DeepSeek V4 Pro 0813$0.152Fits in one request
GLM 5.1$0.324Exceeds the context window

DeepSeek V4 Pro 0813 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 V4 Pro 0813 and GLM 5.1
SpecificationDeepSeek V4 Pro 0813Z.ai GLM 5.1
MakerDeepSeekZ.ai
WeightsOpen weightsOpen weights
ReleasedAugust 12, 2026April 7, 2026
Context window1.05M205K
Input $/Mtok$0.66$1.40
Output $/Mtok$1.98$4.40
Coding index54.644.0
Independent measurements behind it11
Cost per solved task$0.10$0.60
SourceOpenRouter listingOpenRouter listing

Questions

Is DeepSeek V4 Pro 0813 better than GLM 5.1 for coding?

Pick DeepSeek V4 Pro 0813 for coding: it scores 54.6 on the Vibe Coding Index against 44.0 for GLM 5.1. Pick GLM 5.1 when what matters is Terminal-Bench 2.1 (Vals AI) or LMArena Text: coding, where it leads.

Which is cheaper, DeepSeek V4 Pro 0813 or GLM 5.1?

DeepSeek V4 Pro 0813 costs $0.66 per million input tokens and $1.98 per million output tokens; GLM 5.1 costs $1.40 and $4.40. Per solved SWE-bench Verified task, DeepSeek V4 Pro 0813 costs $0.10 and GLM 5.1 $0.60.

Which has the bigger context window, DeepSeek V4 Pro 0813 or GLM 5.1?

DeepSeek V4 Pro 0813, with 1.05M tokens against 205K.

How many benchmarks is the DeepSeek V4 Pro 0813 vs GLM 5.1 comparison based on?

60 benchmarks have a result for both models as of October 21, 2026. DeepSeek V4 Pro 0813's coding index rests on 1 independent benchmark and GLM 5.1's on 1. Where only one model has a result, the page says so instead of naming a winner.

Compare next