LLMBenchmark.ai

Enter opens the highlighted result. Esc closes.

GLM 4.7 vs GLM 5.3

Updated October 21, 2026. 61 benchmarks have a result for both. Coding is ranked on the Vibe Coding Index; every other board is shown as published.

Replace GLM 4.7 Replace GLM 5.3
GLM 4.7Z.ai
Coding
33.6
Overall
22.2
Output $/M
$1.98
1
GLM 5.3Z.ai
Coding
73.1
Overall
44.8
Output $/M
$1.40
60
vs boards won of 61 shared

Decision reading

Pick GLM 5.3 for coding: it scores 73.1 on the Vibe Coding Index against 33.6 for GLM 4.7.

Pick GLM 4.7 when a solved task costs $0.29 against $0.36, or when the benchmark that matters is LiveCodeBench (Vals AI), where it leads.

Model AGLM 4.7Z.ai33.6/1001 independent benchmark$0.29 per solved taskModel BGLM 5.3Z.ai73.1/1002 independent benchmarks$0.36 per solved task
Shared results
61
GLM 4.7 only
19
GLM 5.3 only
39
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
GLM 4.7 33.6 from 1 benchmark, GLM 5.3 73.1 from 2 benchmarks.
Widest category
Agentic work, 23.6 points apart on average over 2 shared benchmarks.
Widest single gap
ProofBench: GLM 4.7 6.0% against GLM 5.3 49.0%, 43.0 points apart on a 0 to 100 scale.
Output price
GLM 4.7 $1.98 against GLM 5.3 $1.40 per million output tokens, 1.4x apart.
Context window
GLM 4.7 takes 205K 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 4.7BarsGLM 5.3Ties
Coding18
Intelligence02
Agentic work02
Reasoning08
Math04
Knowledge09
Web design08
General019
All shared boards160
CodingLike-for-likeFour of these boards feed the Vibe Coding Index18
Coding benchmarks for GLM 4.7 and GLM 5.3
BenchmarkGLM 4.7GLM 5.3Leader
LiveBench codingNot measured82.8%Only GLM 5.3 measured
LMArena WebDev14351621GLM 5.3
AA Coding Index45.374.8GLM 5.3
LiveCodeBench (Vals AI)82.2%80.5%GLM 4.7
SWE-bench Verified (Vals AI)69.4%95.4%GLM 5.3
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)89.4%Not measuredOnly GLM 4.7 measured
SciCodeNot measured59.0%Only GLM 5.3 measured
Terminal-Bench Hard (Artificial Analysis run)31.8%Not measuredOnly GLM 4.7 measured
Terminal-Bench33.4%Not measuredOnly GLM 4.7 measured
FrontierCodeNot measured40.1%Only GLM 5.3 measured
FrontierSWENot measured30.2%Only GLM 5.3 measured
WeirdMLNot measured75.4%Only GLM 5.3 measured
ALE-Bench399.51317.4GLM 5.3
LMArena Text: coding14541497GLM 5.3
LMArena Text: software and IT14531493GLM 5.3
LMArena WebDev: React14201641GLM 5.3
LMArena WebDev: HTML14471595GLM 5.3
Neither model has a result on the other 12 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index02
Intelligence benchmarks for GLM 4.7 and GLM 5.3
BenchmarkGLM 4.7GLM 5.3Leader
AA Intelligence Index22.244.8GLM 5.3
Epoch Capabilities Index143.6156.0GLM 5.3
Agentic workLike-for-likePublic boards, shown as published, kept out of the index02
Agentic work benchmarks for GLM 4.7 and GLM 5.3
BenchmarkGLM 4.7GLM 5.3Leader
AA Agentic IndexNot measured53.1Only GLM 5.3 measured
Design Arena: agents building full-stack apps1050Not measuredOnly GLM 4.7 measured
Design Arena: agents building mobile apps11051194GLM 5.3
Design Arena: agents building Android apps1118Not measuredOnly GLM 4.7 measured
Design Arena: agents building Godot games1058Not measuredOnly GLM 4.7 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 Telecom95.9%Not measuredOnly GLM 4.7 measured
APEX-AgentsNot measured56.6%Only GLM 5.3 measured
Vending-Bench 2$2,377$8,164GLM 5.3
Neither model has a result on the other 19 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index08
Reasoning benchmarks for GLM 4.7 and GLM 5.3
BenchmarkGLM 4.7GLM 5.3Leader
GPQA Diamond80.0%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)85.9%91.7%GLM 5.3
Humanity's Last Exam (Artificial Analysis run)27.4%42.3%GLM 5.3
GPQA Diamond (Epoch AI run)83.3%90.9%GLM 5.3
SimpleBench47.7%Not measuredOnly GLM 4.7 measured
Chess puzzles21.0%37.0%GLM 5.3
Mystery game puzzlesNot measured33.0%Only GLM 5.3 measured
LMArena Text: hard prompts14431490GLM 5.3
LMArena Text: expert14241514GLM 5.3
LMArena Text: hard prompts in English14521496GLM 5.3
Neither model has a result on the other 9 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index04
Math benchmarks for GLM 4.7 and GLM 5.3
BenchmarkGLM 4.7GLM 5.3Leader
AIME93.3%Not measuredOnly GLM 4.7 measured
MGSM (Vals AI)88.2%Not measuredOnly GLM 4.7 measured
AA Math Index95.0Not measuredOnly GLM 4.7 measured
AIME 2025 (Artificial Analysis run)95.0%Not measuredOnly GLM 4.7 measured
OTIS Mock AIME83.3%91.1%GLM 5.3
FrontierMath, tiers 1 to 3Not measured68.8%Only GLM 5.3 measured
FrontierMath, tier 4Not measured29.3%Only GLM 5.3 measured
ProofBench6.0%49.0%GLM 5.3
LMArena Text: math14241495GLM 5.3
LMArena Text: mathematical work14111497GLM 5.3
Neither model has a result on the other 3 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index09
Knowledge benchmarks for GLM 4.7 and GLM 5.3
BenchmarkGLM 4.7GLM 5.3Leader
MMLU Pro82.7%86.8%GLM 5.3
LegalBench (Vals AI)83.4%84.8%GLM 5.3
CaseLaw (Vals AI)54.9%Not measuredOnly GLM 4.7 measured
CorpFin (Vals AI)46.4%Not measuredOnly GLM 4.7 measured
TaxEval (Vals AI)68.8%72.4%GLM 5.3
MedQA (Vals AI)93.7%Not measuredOnly GLM 4.7 measured
MedCode (Vals AI)32.8%42.9%GLM 5.3
MMLU Pro (Artificial Analysis run)85.6%Not measuredOnly GLM 4.7 measured
SimpleQA Verified32.2%41.0%GLM 5.3
LMArena Text: science14571498GLM 5.3
LMArena Text: medicine and healthcare14691481GLM 5.3
LMArena Text: legal and government14441473GLM 5.3
LMArena Text: business and finance14291459GLM 5.3
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 GLM 4.7 and GLM 5.3
BenchmarkGLM 4.7GLM 5.3Leader
Design Arena: websites12261295GLM 5.3
Design Arena: UI components11991314GLM 5.3
Design Arena: game dev11891311GLM 5.3
Design Arena: data viz11951260GLM 5.3
Design Arena: 3D11911332GLM 5.3
Design Arena: all code categories12171304GLM 5.3
Design Arena: SVG11421292GLM 5.3
Design Arena: ASCII art11651213GLM 5.3
LiveBench designNot measured77.3%Only GLM 5.3 measured
GeneralLike-for-likePublic boards, shown as published, kept out of the index019
General benchmarks for GLM 4.7 and GLM 5.3
BenchmarkGLM 4.7GLM 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)68.6%88.8%GLM 5.3
IFBench67.9%Not measuredOnly GLM 4.7 measured
Long context reasoning (AA-LCR)71.0%79.7%GLM 5.3
DTBenchNot measured87.7%Only GLM 5.3 measured
LMCANot measured55.5%Only GLM 5.3 measured
CL-bench15.9%Not measuredOnly GLM 4.7 measured
CL-bench Life10.9%Not measuredOnly GLM 4.7 measured
LMArena Text14351474GLM 5.3
LMArena Text: instruction following14121479GLM 5.3
LMArena Text: creative writing14011458GLM 5.3
LMArena Text: English14531482GLM 5.3
LMArena Text: non-English14171459GLM 5.3
LMArena Text: Chinese14951532GLM 5.3
LMArena Text: German14261488GLM 5.3
LMArena Text: Japanese14391460GLM 5.3
LMArena Text: Korean13991460GLM 5.3
LMArena Text: French14321503GLM 5.3
LMArena Text: Spanish14341462GLM 5.3
LMArena Text: Russian14231469GLM 5.3
LMArena Text: Polish14351466GLM 5.3
LMArena Text: multi-turn14461472GLM 5.3
LMArena Text: longer queries14321483GLM 5.3
LMArena Text: writing and language14111469GLM 5.3
LMArena Text: entertainment and media14001447GLM 5.3
Neither model has a result on the other 19 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for GLM 4.7 and GLM 5.3
BenchmarkGLM 4.7GLM 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 4.7$0.0015Fits in one request
GLM 5.3$0.0008Fits in one request

GLM 5.3 costs less

Repository review

50K input + 3K output tokens

GLM 4.7$0.033Fits in one request
GLM 5.3$0.0081Fits in one request

GLM 5.3 costs less

Agent loop

200K input + 10K output tokens

GLM 4.7$0.128Exceeds 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 4.7 and GLM 5.3
SpecificationZ.ai GLM 4.7Z.ai GLM 5.3
MakerZ.aiZ.ai
WeightsOpen weightsOpen weights
ReleasedDecember 22, 2025August 18, 2026
Context window205K1.05M
Input $/Mtok$0.54$0.079
Output $/Mtok$1.98$1.40
Coding index33.673.1
Independent measurements behind it12
Cost per solved task$0.29$0.36
SourceOpenRouter listingOpenRouter listing

Questions

Is GLM 4.7 better than GLM 5.3 for coding?

Pick GLM 5.3 for coding: it scores 73.1 on the Vibe Coding Index against 33.6 for GLM 4.7. Pick GLM 4.7 when a solved task costs $0.29 against $0.36, or when the benchmark that matters is LiveCodeBench (Vals AI), where it leads.

Which is cheaper, GLM 4.7 or GLM 5.3?

GLM 4.7 costs $0.54 per million input tokens and $1.98 per million output tokens; GLM 5.3 costs $0.079 and $1.40. Per solved SWE-bench Verified task, GLM 4.7 costs $0.29 and GLM 5.3 $0.36.

Which has the bigger context window, GLM 4.7 or GLM 5.3?

GLM 5.3, with 1.05M tokens against 205K.

How many benchmarks is the GLM 4.7 vs GLM 5.3 comparison based on?

61 benchmarks have a result for both models as of October 21, 2026. GLM 4.7'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