LLMBenchmark.ai

Enter opens the highlighted result. Esc closes.

GLM 4.5 vs GLM 5.3

Updated October 21, 2026. 46 benchmarks have a result for both. Coding is ranked on the Vibe Coding Index; every other board is shown as published.

Replace GLM 4.5 Replace GLM 5.3
GLM 4.5Z.ai
Coding
64.2
Overall
12.8
Output $/M
$2.20
1
GLM 5.3Z.ai
Coding
73.1
Overall
44.8
Output $/M
$1.40
45
vs boards won of 46 shared

Decision reading

Pick GLM 5.3 for coding: it scores 73.1 on the Vibe Coding Index against 64.2 for GLM 4.5.

Pick GLM 4.5 when the benchmark that matters is TaxEval (Vals AI), where it leads.

Model AGLM 4.5Z.ai64.2/1001 independent benchmarkNo cost per solved taskModel BGLM 5.3Z.ai73.1/1002 independent benchmarks$0.36 per solved task
Shared results
46
GLM 4.5 only
14
GLM 5.3 only
54
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
GLM 4.5 64.2 from 1 benchmark, GLM 5.3 73.1 from 2 benchmarks.
Widest category
Intelligence, 32.0 points apart on average over 1 shared benchmark.
Widest single gap
WeirdML: GLM 4.5 40.6% against GLM 5.3 75.4%, 34.8 points apart on a 0 to 100 scale.
Output price
GLM 4.5 $2.20 against GLM 5.3 $1.40 per million output tokens, 1.6x apart.
Context window
GLM 4.5 takes 131K tokens, GLM 5.3 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
CategoryGLM 4.5BarsGLM 5.3Ties
Coding05
Intelligence01
Reasoning06
Math02
Knowledge16
Web design07
General018
All shared boards145
CodingLike-for-likeFour of these boards feed the Vibe Coding Index05
Coding benchmarks for GLM 4.5 and GLM 5.3
BenchmarkGLM 4.5GLM 5.3Leader
SWE-bench Verified64.2%Not measuredOnly GLM 4.5 measured
LiveBench codingNot measured82.8%Only GLM 5.3 measured
LMArena WebDevNot measured1621Only GLM 5.3 measured
AA Coding IndexNot measured74.8Only GLM 5.3 measured
LiveCodeBench (Vals AI)67.5%80.5%GLM 5.3
SWE-bench Verified (Vals AI)Not measured95.4%Only GLM 5.3 measured
Terminal-Bench 4 (Vals AI)Not measured38.9%Only GLM 5.3 measured
Vibe Code Bench (Vals AI)Not measured78.1%Only GLM 5.3 measured
Code migration (Vals AI)Not measured44.2%Only GLM 5.3 measured
ProgramBench (Vals AI)Not measured1.5%Only GLM 5.3 measured
IOI (Vals AI)Not measured68.4%Only GLM 5.3 measured
Terminal-Bench 2.1 (Vals AI)Not measured71.5%Only GLM 5.3 measured
Vibe Code Bench 1-100 (Vals AI)Not measured20.0%Only GLM 5.3 measured
SkillsBench (Vals AI)Not measured47.5%Only GLM 5.3 measured
CyberBench (Vals AI)Not measured72.1%Only GLM 5.3 measured
LiveCodeBench (Artificial Analysis run)73.8%Not measuredOnly GLM 4.5 measured
SciCodeNot measured59.0%Only GLM 5.3 measured
Terminal-Bench Hard (Artificial Analysis run)22.0%Not measuredOnly GLM 4.5 measured
FrontierCodeNot measured40.1%Only GLM 5.3 measured
FrontierSWENot measured30.2%Only GLM 5.3 measured
WeirdML40.6%75.4%GLM 5.3
ALE-Bench344.81317.4GLM 5.3
LMArena Text: coding14341497GLM 5.3
LMArena Text: software and IT14401493GLM 5.3
LMArena WebDev: ReactNot measured1641Only GLM 5.3 measured
LMArena WebDev: HTMLNot measured1595Only GLM 5.3 measured
Neither model has a result on the other 12 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index01
Intelligence benchmarks for GLM 4.5 and GLM 5.3
BenchmarkGLM 4.5GLM 5.3Leader
AA Intelligence Index12.844.8GLM 5.3
Epoch Capabilities IndexNot measured156.0Only GLM 5.3 measured
Agentic workDirectional onlyPublic boards, shown as published, kept out of the index17 boards, none shared
Agentic work benchmarks for GLM 4.5 and GLM 5.3
BenchmarkGLM 4.5GLM 5.3Leader
AA Agentic IndexNot measured53.1Only GLM 5.3 measured
Design Arena: agents building mobile appsNot measured1194Only GLM 5.3 measured
Design Arena: HTML slidesNot measured1189Only GLM 5.3 measured
Design Arena: PowerPoint slidesNot measured1255Only GLM 5.3 measured
LiveBench researchNot measured73.2%Only GLM 5.3 measured
LiveBench browserNot measured0.0%Only GLM 5.3 measured
LiveBench long agenticNot measured35.9%Only GLM 5.3 measured
SREBench (Vals AI)Not measured9.2%Only GLM 5.3 measured
Finance Agent (Vals AI)Not measured55.8%Only GLM 5.3 measured
Tax Agent Bench (Vals AI)Not measured39.7%Only GLM 5.3 measured
Legal Research Bench (Vals AI)Not measured49.0%Only GLM 5.3 measured
Harvey's Legal Agent Benchmark (Vals AI)Not measured14.6%Only GLM 5.3 measured
Excel Modeling Benchmark (Vals AI)Not measured56.3%Only GLM 5.3 measured
Terminal-Bench-Science (Vals AI)Not measured5.7%Only GLM 5.3 measured
τ²-Bench Telecom43.0%Not measuredOnly GLM 4.5 measured
APEX-AgentsNot measured56.6%Only GLM 5.3 measured
Vending-Bench 2Not measured$8,164Only GLM 5.3 measured
Neither model has a result on the other 22 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index06
Reasoning benchmarks for GLM 4.5 and GLM 5.3
BenchmarkGLM 4.5GLM 5.3Leader
GPQA Diamond72.2%88.1%GLM 5.3
LiveBench reasoningNot measured85.8%Only GLM 5.3 measured
MysteryMechanism (Vals AI)Not measured23.0%Only GLM 5.3 measured
GPQA Diamond (Artificial Analysis run)78.2%91.7%GLM 5.3
Humanity's Last Exam (Artificial Analysis run)13.0%42.3%GLM 5.3
GPQA Diamond (Epoch AI run)Not measured90.9%Only GLM 5.3 measured
Chess puzzlesNot measured37.0%Only GLM 5.3 measured
Mystery game puzzlesNot measured33.0%Only GLM 5.3 measured
LMArena Text: hard prompts14291490GLM 5.3
LMArena Text: expert14331514GLM 5.3
LMArena Text: hard prompts in English14291496GLM 5.3
Neither model has a result on the other 10 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index02
Math benchmarks for GLM 4.5 and GLM 5.3
BenchmarkGLM 4.5GLM 5.3Leader
AIME86.7%Not measuredOnly GLM 4.5 measured
MATH 500 (Vals AI)94.0%Not measuredOnly GLM 4.5 measured
MGSM (Vals AI)90.8%Not measuredOnly GLM 4.5 measured
AA Math Index73.7Not measuredOnly GLM 4.5 measured
AIME 2025 (Artificial Analysis run)73.7%Not measuredOnly GLM 4.5 measured
OTIS Mock AIMENot measured91.1%Only GLM 5.3 measured
FrontierMath, tiers 1 to 3Not measured68.8%Only GLM 5.3 measured
FrontierMath, tier 4Not measured29.3%Only GLM 5.3 measured
ProofBenchNot measured49.0%Only GLM 5.3 measured
LMArena Text: math14271495GLM 5.3
LMArena Text: mathematical work14301497GLM 5.3
Neither model has a result on the other 2 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index16
Knowledge benchmarks for GLM 4.5 and GLM 5.3
BenchmarkGLM 4.5GLM 5.3Leader
MMLU Pro81.2%86.8%GLM 5.3
LegalBench (Vals AI)75.6%84.8%GLM 5.3
CorpFin (Vals AI)61.0%Not measuredOnly GLM 4.5 measured
TaxEval (Vals AI)72.4%72.4%GLM 4.5
MedQA (Vals AI)90.0%Not measuredOnly GLM 4.5 measured
MedCode (Vals AI)Not measured42.9%Only GLM 5.3 measured
MMLU Pro (Artificial Analysis run)83.5%Not measuredOnly GLM 4.5 measured
SimpleQA VerifiedNot measured41.0%Only GLM 5.3 measured
LMArena Text: science14441498GLM 5.3
LMArena Text: medicine and healthcare14551481GLM 5.3
LMArena Text: legal and government14381473GLM 5.3
LMArena Text: business and finance14201459GLM 5.3
Neither model has a result on the other 5 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index07
Web design benchmarks for GLM 4.5 and GLM 5.3
BenchmarkGLM 4.5GLM 5.3Leader
Design Arena: websites11701295GLM 5.3
Design Arena: UI components11441314GLM 5.3
Design Arena: game dev11451311GLM 5.3
Design Arena: data viz11591260GLM 5.3
Design Arena: 3D11701332GLM 5.3
Design Arena: all code categories11651304GLM 5.3
Design Arena: SVG10961292GLM 5.3
Design Arena: ASCII artNot measured1213Only GLM 5.3 measured
LiveBench designNot measured77.3%Only GLM 5.3 measured
GeneralLike-for-likePublic boards, shown as published, kept out of the index018
General benchmarks for GLM 4.5 and GLM 5.3
BenchmarkGLM 4.5GLM 5.3Leader
Vals IndexNot measured59.7%Only GLM 5.3 measured
LiveBench data analysisNot measured70.2%Only GLM 5.3 measured
Public Benefits Bench (Vals AI)Not measured68.5%Only GLM 5.3 measured
MedScribe (Vals AI)Not measured88.8%Only GLM 5.3 measured
IFBench44.1%Not measuredOnly GLM 4.5 measured
Long context reasoning (AA-LCR)52.7%79.7%GLM 5.3
DTBenchNot measured87.7%Only GLM 5.3 measured
LMCANot measured55.5%Only GLM 5.3 measured
Creative writing (Lech Mazur)73.4%Not measuredOnly GLM 4.5 measured
LMArena Text14301474GLM 5.3
LMArena Text: instruction following14041479GLM 5.3
LMArena Text: creative writing13951458GLM 5.3
LMArena Text: English14371482GLM 5.3
LMArena Text: non-English14171459GLM 5.3
LMArena Text: Chinese14651532GLM 5.3
LMArena Text: German14071488GLM 5.3
LMArena Text: Japanese14151460GLM 5.3
LMArena Text: Korean13801460GLM 5.3
LMArena Text: French14181503GLM 5.3
LMArena Text: Spanish14541462GLM 5.3
LMArena Text: Russian14131469GLM 5.3
LMArena Text: Polish14221466GLM 5.3
LMArena Text: multi-turn14151472GLM 5.3
LMArena Text: longer queries14121483GLM 5.3
LMArena Text: writing and language14031469GLM 5.3
LMArena Text: entertainment and media14041447GLM 5.3
Neither model has a result on the other 20 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for GLM 4.5 and GLM 5.3
BenchmarkGLM 4.5GLM 5.3Leader
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

GLM 4.5$0.0017Fits in one request
GLM 5.3$0.0008Fits in one request

GLM 5.3 costs less

Repository review

50K input + 3K output tokens

GLM 4.5$0.037Fits in one request
GLM 5.3$0.0081Fits in one request

GLM 5.3 costs less

Agent loop

200K input + 10K output tokens

GLM 4.5$0.142Exceeds the context window
GLM 5.3$0.030Fits in one request

GLM 5.3 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 GLM 4.5 and GLM 5.3
SpecificationZ.ai GLM 4.5Z.ai GLM 5.3
MakerZ.aiZ.ai
WeightsOpen weightsOpen weights
ReleasedJuly 25, 2025August 18, 2026
Context window131K1.05M
Input $/Mtok$0.60$0.079
Output $/Mtok$2.20$1.40
Coding index64.273.1
Independent measurements behind it12
Cost per solved taskNot computable$0.36
SourceOpenRouter listingOpenRouter listing

Questions

Is GLM 4.5 better than GLM 5.3 for coding?

Pick GLM 5.3 for coding: it scores 73.1 on the Vibe Coding Index against 64.2 for GLM 4.5. Pick GLM 4.5 when the benchmark that matters is TaxEval (Vals AI), where it leads.

Which is cheaper, GLM 4.5 or GLM 5.3?

GLM 4.5 costs $0.60 per million input tokens and $2.20 per million output tokens; GLM 5.3 costs $0.079 and $1.40. Cost per solved task is known for GLM 5.3 only.

Which has the bigger context window, GLM 4.5 or GLM 5.3?

GLM 5.3, with 1.05M tokens against 131K.

How many benchmarks is the GLM 4.5 vs GLM 5.3 comparison based on?

46 benchmarks have a result for both models as of October 21, 2026. GLM 4.5's coding index rests on 1 independent benchmark and GLM 5.3's on 2. Where only one model has a result, the page says so instead of naming a winner.

Compare next