Enter opens the highlighted result. Esc closes.

GLM 5.1 vs GLM 5.3

GLM 5.3 leads GLM 5.1 on coding, 70.6 to 49.2 on the Vibe Coding Index, at about the same output price ($4.40 vs $4.40 per million). Updated October 22, 2026; 75 benchmarks have a result for both.

Replace GLM 5.1 Replace GLM 5.3
GLM 5.1Z.ai
Coding
49.2
Overall
57.5
Output $/M
$4.40
7
GLM 5.3Z.ai
Coding
70.6
Overall
67.9
Output $/M
$4.40
68
vs boards won of 75 shared

GLM 5.1 or GLM 5.3: which to pick

Pick GLM 5.3 for coding: it scores 70.6 on the Vibe Coding Index against 49.2 for GLM 5.1.

Pick GLM 5.1 for work measured by LiveCodeBench (Vals AI) or OTIS Mock AIME, where it leads.

Model AGLM 5.1Z.ai49.2/1003 independent benchmarks$0.60 per solved taskModel BGLM 5.3Z.ai70.6/1006 independent benchmarks$0.36 per solved task
Shared results
75
GLM 5.1 only
14
GLM 5.3 only
25
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
GLM 5.1 49.2 from 3 benchmarks, GLM 5.3 70.6 from 6 benchmarks.
Widest category
Coding, 14.5 points apart on average over 15 shared benchmarks.
Widest single gap
Vibe Code Bench (Vals AI): GLM 5.1 31.5% against GLM 5.3 78.1%, 46.7 points apart on a 0 to 100 scale.
Output price
Both cost $4.40 per million output tokens.
Context window
GLM 5.1 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 5.1BarsGLM 5.3Ties
Coding114
Intelligence02
Agentic work08
Reasoning08
Math14
Knowledge18
Web design26
General218
All shared boards768
CodingLike-for-likeSix of these boards feed the Vibe Coding Index114
Coding benchmarks for GLM 5.1 and GLM 5.3
BenchmarkGLM 5.1GLM 5.3Leader
LiveBench codingNot measured82.8%Only GLM 5.3 measured
LMArena WebDev15081621GLM 5.3
AA Coding Index55.874.8GLM 5.3
LiveCodeBench (Vals AI)81.4%80.5%GLM 5.1
SWE-bench Verified (Vals AI)76.4%95.4%GLM 5.3
Terminal-Bench 4 (Vals AI)Not measured38.9%Only GLM 5.3 measured
Vibe Code Bench (Vals AI)31.5%78.1%GLM 5.3
Code migration (Vals AI)25.8%44.2%GLM 5.3
ProgramBench (Vals AI)0.0%1.5%GLM 5.3
IOI (Vals AI)Not measured68.4%Only GLM 5.3 measured
Terminal-Bench 2.1 (Vals AI)56.9%71.5%GLM 5.3
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
SciCode44.8%59.0%GLM 5.3
Terminal-Bench Hard (Artificial Analysis run)43.2%Not measuredOnly GLM 5.1 measured
FrontierCodeNot measured40.1%Only GLM 5.3 measured
FrontierSWENot measured30.2%Only GLM 5.3 measured
WeirdML57.1%75.4%GLM 5.3
ALE-Bench887.11317.4GLM 5.3
LMArena Text: coding14851497GLM 5.3
LMArena Text: software and IT14841493GLM 5.3
LMArena WebDev: React15051641GLM 5.3
LMArena WebDev: HTML15081595GLM 5.3
Neither model has a result on the other 15 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index02
Intelligence benchmarks for GLM 5.1 and GLM 5.3
BenchmarkGLM 5.1GLM 5.3Leader
AA Intelligence Index26.144.8GLM 5.3
Epoch Capabilities Index149.8156.0GLM 5.3
Agentic workLike-for-likePublic boards, shown as published, kept out of the index08
Agentic work benchmarks for GLM 5.1 and GLM 5.3
BenchmarkGLM 5.1GLM 5.3Leader
AA Agentic Index23.953.1GLM 5.3
Design Arena: agents building web apps1175Not measuredOnly GLM 5.1 measured
Design Arena: agents building full-stack apps1167Not measuredOnly GLM 5.1 measured
Design Arena: agents building mobile apps11621194GLM 5.3
Design Arena: agents building Android apps1182Not measuredOnly GLM 5.1 measured
Design Arena: agents building Godot games1053Not measuredOnly GLM 5.1 measured
Design Arena: agents building games1164Not measuredOnly GLM 5.1 measured
Design Arena: HTML slides11201189GLM 5.3
Design Arena: PowerPoint slides11631230GLM 5.3
Design Arena: spreadsheets1315Not measuredOnly GLM 5.1 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)44.8%55.8%GLM 5.3
Tax Agent Bench (Vals AI)Not measured39.7%Only GLM 5.3 measured
Legal Research Bench (Vals AI)27.9%49.0%GLM 5.3
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 Telecom97.7%Not measuredOnly GLM 5.1 measured
APEX-Agents40.9%56.6%GLM 5.3
Vending-Bench 2$5,634$8,164GLM 5.3
Neither model has a result on the other 18 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index08
Reasoning benchmarks for GLM 5.1 and GLM 5.3
BenchmarkGLM 5.1GLM 5.3Leader
GPQA Diamond84.5%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)86.8%91.7%GLM 5.3
Humanity's Last Exam (Artificial Analysis run)30.1%42.3%GLM 5.3
GPQA Diamond (Epoch AI run)89.9%90.9%GLM 5.3
SimpleBench55.1%Not measuredOnly GLM 5.1 measured
Chess puzzles20.0%37.0%GLM 5.3
Mystery game puzzlesNot measured33.0%Only GLM 5.3 measured
LMArena Text: hard prompts14731490GLM 5.3
LMArena Text: expert14781514GLM 5.3
LMArena Text: hard prompts in English14821496GLM 5.3
Neither model has a result on the other 10 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index14
Math benchmarks for GLM 5.1 and GLM 5.3
BenchmarkGLM 5.1GLM 5.3Leader
AIME91.9%Not measuredOnly GLM 5.1 measured
OTIS Mock AIME93.3%91.1%GLM 5.1
FrontierMath, tiers 1 to 336.8%68.8%GLM 5.3
FrontierMath, tier 4Not measured29.3%Only GLM 5.3 measured
ProofBench22.2%49.0%GLM 5.3
LMArena Text: math14731495GLM 5.3
LMArena Text: mathematical work14701497GLM 5.3
Neither model has a result on the other 6 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index18
Knowledge benchmarks for GLM 5.1 and GLM 5.3
BenchmarkGLM 5.1GLM 5.3Leader
MMLU Pro86.9%86.8%GLM 5.1
LegalBench (Vals AI)84.4%84.8%GLM 5.3
CaseLaw (Vals AI)51.5%Not measuredOnly GLM 5.1 measured
CorpFin (Vals AI)64.5%Not measuredOnly GLM 5.1 measured
TaxEval (Vals AI)71.2%72.4%GLM 5.3
MedCode (Vals AI)41.6%42.9%GLM 5.3
SimpleQA Verified34.0%41.0%GLM 5.3
LMArena Text: science14781498GLM 5.3
LMArena Text: medicine and healthcare14741481GLM 5.3
LMArena Text: legal and government14651473GLM 5.3
LMArena Text: business and finance14521459GLM 5.3
Neither model has a result on the other 7 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index26
Web design benchmarks for GLM 5.1 and GLM 5.3
BenchmarkGLM 5.1GLM 5.3Leader
Design Arena: websites12901295GLM 5.3
Design Arena: UI components12921314GLM 5.3
Design Arena: game dev12651311GLM 5.3
Design Arena: data viz13661260GLM 5.1
Design Arena: 3D13361332GLM 5.1
Design Arena: all code categories12681304GLM 5.3
Design Arena: SVG12181294GLM 5.3
Design Arena: ASCII art11411208GLM 5.3
LiveBench designNot measured77.3%Only GLM 5.3 measured
GeneralLike-for-likePublic boards, shown as published, kept out of the index218
General benchmarks for GLM 5.1 and GLM 5.3
BenchmarkGLM 5.1GLM 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)61.8%68.5%GLM 5.3
MedScribe (Vals AI)72.3%88.8%GLM 5.3
IFBench76.3%Not measuredOnly GLM 5.1 measured
Long context reasoning (AA-LCR)73.7%79.7%GLM 5.3
DTBenchNot measured87.7%Only GLM 5.3 measured
LMCANot measured55.5%Only GLM 5.3 measured
ExploitBench18.1%Not measuredOnly GLM 5.1 measured
LMArena Text14631474GLM 5.3
LMArena Text: instruction following14531479GLM 5.3
LMArena Text: creative writing14531458GLM 5.3
LMArena Text: English14761482GLM 5.3
LMArena Text: non-English14481459GLM 5.3
LMArena Text: Chinese15151532GLM 5.3
LMArena Text: German14671488GLM 5.3
LMArena Text: Japanese14351460GLM 5.3
LMArena Text: Korean14241460GLM 5.3
LMArena Text: French14701503GLM 5.3
LMArena Text: Spanish14681462GLM 5.1
LMArena Text: Russian14541469GLM 5.3
LMArena Text: Polish14491466GLM 5.3
LMArena Text: multi-turn14731472GLM 5.1
LMArena Text: longer queries14671483GLM 5.3
LMArena Text: writing and language14541469GLM 5.3
LMArena Text: entertainment and media14421447GLM 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 5.1 and GLM 5.3
BenchmarkGLM 5.1GLM 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 5.1$0.0036Fits in one request
GLM 5.3$0.0036Fits in one request

Same cost

Repository review

50K input + 3K output tokens

GLM 5.1$0.083Fits in one request
GLM 5.3$0.083Fits in one request

Same cost

Agent loop

200K input + 10K output tokens

GLM 5.1$0.324Exceeds the context window
GLM 5.3$0.324Fits in one request

Same cost

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 5.1 and GLM 5.3
SpecificationZ.ai GLM 5.1Z.ai GLM 5.3
MakerZ.aiZ.ai
WeightsOpen weightsOpen weights
ReleasedApril 7, 2026August 18, 2026
Context window205K1.05M
Input $/Mtok$1.40$1.40
Output $/Mtok$4.40$4.40
Coding index49.270.6
Independent measurements behind it36
Cost per solved task$0.60$0.36
SourceOpenRouter listingOpenRouter listing

Questions

Is GLM 5.1 better than GLM 5.3 for coding?

Pick GLM 5.3 for coding: it scores 70.6 on the Vibe Coding Index against 49.2 for GLM 5.1. Pick GLM 5.1 for work measured by LiveCodeBench (Vals AI) or OTIS Mock AIME, where it leads.

Which is cheaper, GLM 5.1 or GLM 5.3?

GLM 5.1 costs $1.40 per million input tokens and $4.40 per million output tokens; GLM 5.3 costs $1.40 and $4.40. Per solved SWE-bench Verified task, GLM 5.1 costs $0.60 and GLM 5.3 $0.36.

Which has the bigger context window, GLM 5.1 or GLM 5.3?

GLM 5.3, with 1.05M tokens against 205K.

How many benchmarks is the GLM 5.1 vs GLM 5.3 comparison based on?

75 benchmarks have a result for both models as of October 22, 2026. GLM 5.1's coding index rests on 3 independent benchmarks and GLM 5.3's on 6. Where only one model has a result, the page says so instead of naming a winner.

Compare next

Embed this comparison on your site

Free under CC BY 4.0, while the source link stays visible. Preview the frame · Terms and other embeds

<iframe src="https://llmbenchmark.ai/embed/compare/glm-5-1-vs-glm-5-3" title="GLM 5.1 vs GLM 5.3, LLMBenchmark.ai" width="100%" height="360" style="border:0;max-width:100%" loading="lazy"></iframe>
<p style="font:13px/1.4 sans-serif;margin:6px 0 0">Source: <a href="https://llmbenchmark.ai/compare/glm-5-1-vs-glm-5-3">LLMBenchmark.ai, GLM 5.1 vs GLM 5.3</a></p>