Enter opens the highlighted result. Esc closes.

GLM 5.3 vs Kimi K3

GLM 5.3 leads Kimi K3 on coding, 70.6 to 62.3 on the Vibe Coding Index, and GLM 5.3 costs 3.4x less per output token ($4.40 vs $15.00 per million). Updated October 21, 2026; 96 benchmarks have a result for both.

Replace GLM 5.3 Replace Kimi K3
GLM 5.3Z.ai
Coding
70.6
Overall
67.9
Output $/M
$4.40
28
Kimi K3Moonshot AI
Coding
62.3
Overall
66.5
Output $/M
$15.00
67
vs boards won of 96 shared, 1 tied

GLM 5.3 or Kimi K3: which to pick

Pick GLM 5.3 for coding: it scores 70.6 on the Vibe Coding Index against 62.3 for Kimi K3.

Pick Kimi K3 for work measured by LMArena WebDev or AA Coding Index, where it leads.

Model AGLM 5.3Z.ai70.6/1006 independent benchmarks$0.36 per solved taskModel BKimi K3Moonshot AI62.3/1006 independent benchmarks$0.81 per solved task
Shared results
96
GLM 5.3 only
4
Kimi K3 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
GLM 5.3 70.6 from 6 benchmarks, Kimi K3 62.3 from 6 benchmarks.
Widest category
Math, 9.8 points apart on average over 6 shared benchmarks.
Widest single gap
ProofBench: GLM 5.3 49.0% against Kimi K3 87.0%, 38.0 points apart on a 0 to 100 scale.
Output price
GLM 5.3 $4.40 against Kimi K3 $15.00 per million output tokens, 3.4x apart.
Context window
Both take 1.05M tokens.

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 5.3BarsKimi K3Ties
Coding715
Intelligence11
Agentic work87
Reasoning181 tie
Math15
Knowledge09
Web design17
General915
All shared boards28671 tie
CodingLike-for-likeSix of these boards feed the Vibe Coding Index715
Coding benchmarks for GLM 5.3 and Kimi K3
BenchmarkGLM 5.3Kimi K3Leader
LiveBench coding82.8%59.3%GLM 5.3
LMArena WebDev16211655Kimi K3
AA Coding Index74.876.2Kimi K3
LiveCodeBench (Vals AI)80.5%87.2%Kimi K3
SWE-bench Verified (Vals AI)95.4%93.4%GLM 5.3
Terminal-Bench 4 (Vals AI)38.9%17.2%GLM 5.3
Vibe Code Bench (Vals AI)78.1%85.0%Kimi K3
Code migration (Vals AI)44.2%16.1%GLM 5.3
ProgramBench (Vals AI)1.5%2.0%Kimi K3
IOI (Vals AI)68.4%48.9%GLM 5.3
Terminal-Bench 2.1 (Vals AI)71.5%80.9%Kimi K3
SWE-bench Multimodal (Vals AI)Not measured44.2%Only Kimi K3 measured
Vibe Code Bench 1-100 (Vals AI)20.0%18.2%GLM 5.3
SkillsBench (Vals AI)47.5%Not measuredOnly GLM 5.3 measured
CyberBench (Vals AI)72.1%75.2%Kimi K3
SciCode59.0%59.5%Kimi K3
FrontierCode40.1%44.2%Kimi K3
FrontierSWE30.2%25.9%GLM 5.3
WeirdML75.4%82.6%Kimi K3
ALE-Bench1317.41524.5Kimi K3
LMArena Text: coding14971508Kimi K3
LMArena Text: software and IT14931499Kimi K3
LMArena WebDev: React16411670Kimi K3
LMArena WebDev: HTML15951644Kimi K3
LMArena WebDev: from an imageNot measured1579Only Kimi K3 measured
Neither model has a result on the other 14 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index11
Intelligence benchmarks for GLM 5.3 and Kimi K3
BenchmarkGLM 5.3Kimi K3Leader
AA Intelligence Index44.843.6GLM 5.3
Epoch Capabilities Index156.0157.9Kimi K3
Agentic workLike-for-likePublic boards, shown as published, kept out of the index87
Agentic work benchmarks for GLM 5.3 and Kimi K3
BenchmarkGLM 5.3Kimi K3Leader
AA Agentic Index53.150.0GLM 5.3
Design Arena: agents building web appsNot measured1295Only Kimi K3 measured
Design Arena: agents building full-stack appsNot measured1315Only Kimi K3 measured
Design Arena: agents building mobile apps11941269Kimi K3
Design Arena: agents building Android appsNot measured1250Only Kimi K3 measured
Design Arena: agents building Godot gamesNot measured1223Only Kimi K3 measured
Design Arena: agents building gamesNot measured1252Only Kimi K3 measured
Design Arena: HTML slides11891256Kimi K3
Design Arena: PowerPoint slides12541273Kimi K3
Design Arena: spreadsheetsNot measured1058Only Kimi K3 measured
LiveBench research73.2%79.0%Kimi K3
LiveBench browser0.0%16.0%Kimi K3
LiveBench long agentic35.9%42.8%Kimi K3
SREBench (Vals AI)9.2%Not measuredOnly GLM 5.3 measured
Finance Agent (Vals AI)55.8%53.1%GLM 5.3
Tax Agent Bench (Vals AI)39.7%33.4%GLM 5.3
Legal Research Bench (Vals AI)49.0%46.1%GLM 5.3
Harvey's Legal Agent Benchmark (Vals AI)14.6%12.9%GLM 5.3
Excel Modeling Benchmark (Vals AI)56.3%66.7%Kimi K3
BioMysteryBench (Vals AI)Not measured72.6%Only Kimi K3 measured
Terminal-Bench-Science (Vals AI)5.7%1.4%GLM 5.3
Time Horizon Index (Vals AI)Not measured10.5%Only Kimi K3 measured
APEX-Agents56.6%47.8%GLM 5.3
PostTrainBenchNot measured32.0%Only Kimi K3 measured
Vending-Bench 2$8,164$5,165GLM 5.3
Neither model has a result on the other 16 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index18
Reasoning benchmarks for GLM 5.3 and Kimi K3
BenchmarkGLM 5.3Kimi K3Leader
GPQA Diamond88.1%92.9%Kimi K3
LiveBench reasoning85.8%90.7%Kimi K3
MysteryMechanism (Vals AI)23.0%Not measuredOnly GLM 5.3 measured
MMMU Pro (Vals AI)Not measured88.2%Only Kimi K3 measured
GPQA Diamond (Artificial Analysis run)91.7%93.5%Kimi K3
Humanity's Last Exam (Artificial Analysis run)42.3%46.9%Kimi K3
GPQA Diamond (Epoch AI run)90.9%93.1%Kimi K3
ARC-AGINot measured94.5%Only Kimi K3 measured
ARC-AGI-2Not measured60.4%Only Kimi K3 measured
SimpleBenchNot measured60.7%Only Kimi K3 measured
Chess puzzles37.0%37.0%Tie
Mystery game puzzles33.0%26.0%GLM 5.3
LMArena Text: hard prompts14901495Kimi K3
LMArena Text: expert15141520Kimi K3
LMArena Text: hard prompts in English14961496Kimi K3
Neither model has a result on the other 7 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index15
Math benchmarks for GLM 5.3 and Kimi K3
BenchmarkGLM 5.3Kimi K3Leader
OTIS Mock AIME91.1%97.2%Kimi K3
FrontierMath, tiers 1 to 368.8%72.2%Kimi K3
FrontierMath, tier 429.3%39.0%Kimi K3
ProofBench49.0%87.0%Kimi K3
LMArena Text: math14951494GLM 5.3
LMArena Text: mathematical work14971505Kimi K3
Neither model has a result on the other 7 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index09
Knowledge benchmarks for GLM 5.3 and Kimi K3
BenchmarkGLM 5.3Kimi K3Leader
MMLU Pro86.8%88.0%Kimi K3
LegalBench (Vals AI)84.8%86.2%Kimi K3
CorpFin (Vals AI)Not measured71.6%Only Kimi K3 measured
TaxEval (Vals AI)72.4%75.7%Kimi K3
MortgageTax (Vals AI)Not measured66.3%Only Kimi K3 measured
MedCode (Vals AI)42.9%49.4%Kimi K3
SimpleQA Verified41.0%50.6%Kimi K3
LMArena Text: science14981508Kimi K3
LMArena Text: medicine and healthcare14811489Kimi K3
LMArena Text: legal and government14731503Kimi K3
LMArena Text: business and finance14591470Kimi K3
Neither model has a result on the other 7 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index17
Web design benchmarks for GLM 5.3 and Kimi K3
BenchmarkGLM 5.3Kimi K3Leader
Design Arena: websites12951335Kimi K3
Design Arena: UI components13141344Kimi K3
Design Arena: game dev13111359Kimi K3
Design Arena: data viz12591351Kimi K3
Design Arena: 3D13321388Kimi K3
Design Arena: all code categories13041360Kimi K3
Design Arena: SVG12931302Kimi K3
Design Arena: ASCII art1213Not measuredOnly GLM 5.3 measured
LiveBench design77.3%77.2%GLM 5.3
GeneralLike-for-likePublic boards, shown as published, kept out of the index915
General benchmarks for GLM 5.3 and Kimi K3
BenchmarkGLM 5.3Kimi K3Leader
Vals Index59.7%53.0%GLM 5.3
LiveBench data analysis70.2%78.7%Kimi K3
Public Benefits Bench (Vals AI)68.5%68.2%GLM 5.3
SAGE (Vals AI)Not measured52.8%Only Kimi K3 measured
MedScribe (Vals AI)88.8%88.0%GLM 5.3
Vals Multimodal IndexNot measured73.4%Only Kimi K3 measured
Long context reasoning (AA-LCR)79.7%88.7%Kimi K3
DTBench87.7%91.2%Kimi K3
LMCA55.5%52.7%GLM 5.3
Surface Evolver BenchNot measured95.0%Only Kimi K3 measured
Furniture assemblyNot measured34.2%Only Kimi K3 measured
LMArena Text14741476Kimi K3
LMArena Text: instruction following14791485Kimi K3
LMArena Text: creative writing14581457GLM 5.3
LMArena Text: English14821481GLM 5.3
LMArena Text: non-English14591466Kimi K3
LMArena Text: Chinese15321529GLM 5.3
LMArena Text: German14881490Kimi K3
LMArena Text: Japanese14601494Kimi K3
LMArena Text: Korean14601460Kimi K3
LMArena Text: French15031492GLM 5.3
LMArena Text: Spanish14621472Kimi K3
LMArena Text: Russian14691485Kimi K3
LMArena Text: Polish14661458GLM 5.3
LMArena Text: multi-turn14721489Kimi K3
LMArena Text: longer queries14831494Kimi K3
LMArena Text: writing and language14691471Kimi K3
LMArena Text: entertainment and media14471453Kimi K3
Neither model has a result on the other 18 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for GLM 5.3 and Kimi K3
BenchmarkGLM 5.3Kimi K3Leader
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 5.3$0.0036Fits in one request
Kimi K3$0.011Fits in one request

GLM 5.3 costs less

Repository review

50K input + 3K output tokens

GLM 5.3$0.083Fits in one request
Kimi K3$0.195Fits in one request

GLM 5.3 costs less

Agent loop

200K input + 10K output tokens

GLM 5.3$0.324Fits in one request
Kimi K3$0.750Fits 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 5.3 and Kimi K3
SpecificationZ.ai GLM 5.3MoonshotAI Kimi K3
MakerZ.aiMoonshot AI
WeightsOpen weightsOpen weights
ReleasedAugust 18, 2026July 16, 2026
Context window1.05M1.05M
Input $/Mtok$1.40$3.00
Output $/Mtok$4.40$15.00
Coding index70.662.3
Independent measurements behind it66
Cost per solved task$0.36$0.81
SourceOpenRouter listingOpenRouter listing

Questions

Is GLM 5.3 better than Kimi K3 for coding?

Pick GLM 5.3 for coding: it scores 70.6 on the Vibe Coding Index against 62.3 for Kimi K3. Pick Kimi K3 for work measured by LMArena WebDev or AA Coding Index, where it leads.

Which is cheaper, GLM 5.3 or Kimi K3?

GLM 5.3 costs $1.40 per million input tokens and $4.40 per million output tokens; Kimi K3 costs $3.00 and $15.00. Per solved SWE-bench Verified task, GLM 5.3 costs $0.36 and Kimi K3 $0.81.

Which has the bigger context window, GLM 5.3 or Kimi K3?

Neither: both take 1.05M tokens.

How many benchmarks is the GLM 5.3 vs Kimi K3 comparison based on?

96 benchmarks have a result for both models as of October 21, 2026. GLM 5.3's coding index rests on 6 independent benchmarks and Kimi K3'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-5-3-vs-kimi-k3" title="GLM 5.3 vs Kimi K3, 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-5-3-vs-kimi-k3">LLMBenchmark.ai, GLM 5.3 vs Kimi K3</a></p>