Enter opens the highlighted result. Esc closes.

GLM 4.6 vs Grok 4.20

Grok 4.20 leads GLM 4.6 on coding, 33.3 to 8.8 on the Vibe Coding Index, at about the same output price ($2.50 vs $2.20 per million). Updated October 21, 2026; 28 benchmarks have a result for both.

Replace GLM 4.6 Replace Grok 4.20
GLM 4.6Z.ai
Coding
8.8
Overall
18.5
Output $/M
$2.20
4
Grok 4.20xAI
Coding
33.3
Overall
25.7
Output $/M
$2.50
24
vs boards won of 28 shared

GLM 4.6 or Grok 4.20: which to pick

Pick Grok 4.20 for coding: it scores 33.3 on the Vibe Coding Index against 8.8 for GLM 4.6.

Pick GLM 4.6 for output tokens 1.1x cheaper ($2.20 vs $2.50 per million) and work measured by Design Arena: agents building mobile apps or Design Arena: agents building Android apps, where it leads.

Model AGLM 4.6Z.ai8.8/1002 independent benchmarksNo cost per solved taskModel BGrok 4.20xAI33.3/1002 independent benchmarks$0.72 per solved task
Shared results
28
GLM 4.6 only
38
Grok 4.20 only
26
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 4.6 8.8 from 2 benchmarks, Grok 4.20 33.3 from 2 benchmarks.
Widest category
General, 26.4 points apart on average over 2 shared benchmarks.
Widest single gap
IFBench: GLM 4.6 43.4% against Grok 4.20 81.2%, 37.8 points apart on a 0 to 100 scale.
Output price
GLM 4.6 $2.20 against Grok 4.20 $2.50 per million output tokens, 1.1x apart.
Context window
GLM 4.6 takes 205K tokens, Grok 4.20 2M.

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 4.6BarsGrok 4.20Ties
Coding04
Intelligence01
Agentic work32
Reasoning03
Math01
Knowledge14
Web design07
General02
All shared boards424
CodingLike-for-likeSix of these boards feed the Vibe Coding Index04
Coding benchmarks for GLM 4.6 and Grok 4.20
BenchmarkGLM 4.6Grok 4.20Leader
SWE-bench Verified68.2%Not measuredOnly GLM 4.6 measured
LMArena WebDev1341Not measuredOnly GLM 4.6 measured
AA Coding Index45.8Not measuredOnly GLM 4.6 measured
LiveCodeBench (Vals AI)81.0%84.3%Grok 4.20
SWE-bench Verified (Vals AI)Not measured72.2%Only Grok 4.20 measured
Vibe Code Bench (Vals AI)3.1%4.1%Grok 4.20
Code migration (Vals AI)Not measured0.3%Only Grok 4.20 measured
Terminal-Bench 2.1 (Vals AI)Not measured44.2%Only Grok 4.20 measured
LiveCodeBench (Artificial Analysis run)69.5%Not measuredOnly GLM 4.6 measured
Terminal-Bench Hard (Artificial Analysis run)28.8%37.9%Grok 4.20
Terminal-Bench24.5%57.3%Grok 4.20
WeirdMLNot measured52.3%Only Grok 4.20 measured
ALE-Bench340.8Not measuredOnly GLM 4.6 measured
LMArena Text: coding1449Not measuredOnly GLM 4.6 measured
LMArena Text: software and IT1455Not measuredOnly GLM 4.6 measured
LMArena WebDev: HTML1350Not measuredOnly GLM 4.6 measured
Neither model has a result on the other 22 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index01
Intelligence benchmarks for GLM 4.6 and Grok 4.20
BenchmarkGLM 4.6Grok 4.20Leader
AA Intelligence Index18.525.7Grok 4.20
Epoch Capabilities IndexNot measured152.1Only Grok 4.20 measured
Agentic workLike-for-likePublic boards, shown as published, kept out of the index32
Agentic work benchmarks for GLM 4.6 and Grok 4.20
BenchmarkGLM 4.6Grok 4.20Leader
Design Arena: agents building web appsNot measured1144Only Grok 4.20 measured
Design Arena: agents building full-stack apps10291054Grok 4.20
Design Arena: agents building mobile apps10961090GLM 4.6
Design Arena: agents building Android apps10841052GLM 4.6
Design Arena: agents building Godot games11801089GLM 4.6
Design Arena: HTML slidesNot measured1180Only Grok 4.20 measured
Finance Agent (Vals AI)Not measured28.5%Only Grok 4.20 measured
Tax Agent Bench (Vals AI)Not measured7.0%Only Grok 4.20 measured
Legal Research Bench (Vals AI)Not measured13.9%Only Grok 4.20 measured
Excel Modeling Benchmark (Vals AI)Not measured11.7%Only Grok 4.20 measured
τ²-Bench Telecom76.9%93.0%Grok 4.20
Vending-Bench 2Not measured$4,663Only Grok 4.20 measured
Neither model has a result on the other 27 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index03
Reasoning benchmarks for GLM 4.6 and Grok 4.20
BenchmarkGLM 4.6Grok 4.20Leader
GPQA Diamond74.5%88.6%Grok 4.20
MMMU Pro (Vals AI)Not measured83.5%Only Grok 4.20 measured
GPQA Diamond (Artificial Analysis run)78.0%91.1%Grok 4.20
Humanity's Last Exam (Artificial Analysis run)14.5%34.5%Grok 4.20
ARC-AGINot measured89.5%Only Grok 4.20 measured
ARC-AGI-2Not measured65.1%Only Grok 4.20 measured
LMArena Text: hard prompts1440Not measuredOnly GLM 4.6 measured
LMArena Text: expert1432Not measuredOnly GLM 4.6 measured
LMArena Text: hard prompts in English1445Not measuredOnly GLM 4.6 measured
Neither model has a result on the other 12 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index01
Math benchmarks for GLM 4.6 and Grok 4.20
BenchmarkGLM 4.6Grok 4.20Leader
AIME92.7%96.5%Grok 4.20
MGSM (Vals AI)89.8%Not measuredOnly GLM 4.6 measured
AA Math Index86.0Not measuredOnly GLM 4.6 measured
AIME 2025 (Artificial Analysis run)86.0%Not measuredOnly GLM 4.6 measured
LMArena Text: math1432Not measuredOnly GLM 4.6 measured
LMArena Text: mathematical work1447Not measuredOnly GLM 4.6 measured
Neither model has a result on the other 7 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index14
Knowledge benchmarks for GLM 4.6 and Grok 4.20
BenchmarkGLM 4.6Grok 4.20Leader
MMLU Pro82.2%86.3%Grok 4.20
LegalBench (Vals AI)79.6%77.7%GLM 4.6
CaseLaw (Vals AI)Not measured54.4%Only Grok 4.20 measured
CorpFin (Vals AI)56.8%63.7%Grok 4.20
TaxEval (Vals AI)66.2%74.1%Grok 4.20
MortgageTax (Vals AI)Not measured45.4%Only Grok 4.20 measured
MedQA (Vals AI)92.2%94.5%Grok 4.20
MedCode (Vals AI)Not measured32.2%Only Grok 4.20 measured
MMLU Pro (Artificial Analysis run)82.9%Not measuredOnly GLM 4.6 measured
LMArena Text: science1456Not measuredOnly GLM 4.6 measured
LMArena Text: medicine and healthcare1455Not measuredOnly GLM 4.6 measured
LMArena Text: legal and government1448Not measuredOnly GLM 4.6 measured
LMArena Text: business and finance1416Not measuredOnly GLM 4.6 measured
Neither model has a result on the other 4 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index07
Web design benchmarks for GLM 4.6 and Grok 4.20
BenchmarkGLM 4.6Grok 4.20Leader
Design Arena: websites11741229Grok 4.20
Design Arena: UI components11481199Grok 4.20
Design Arena: game dev11511192Grok 4.20
Design Arena: data viz11611217Grok 4.20
Design Arena: 3D11291193Grok 4.20
Design Arena: all code categories11661221Grok 4.20
Design Arena: SVG11091164Grok 4.20
Design Arena: ASCII artNot measured1174Only Grok 4.20 measured
Neither model has a result on the other 1 board here
GeneralLike-for-likePublic boards, shown as published, kept out of the index02
General benchmarks for GLM 4.6 and Grok 4.20
BenchmarkGLM 4.6Grok 4.20Leader
SAGE (Vals AI)Not measured38.2%Only Grok 4.20 measured
MedScribe (Vals AI)Not measured63.4%Only Grok 4.20 measured
Vals Multimodal IndexNot measured39.1%Only Grok 4.20 measured
IFBench43.4%81.2%Grok 4.20
Long context reasoning (AA-LCR)54.0%69.0%Grok 4.20
DTBenchNot measured90.1%Only Grok 4.20 measured
LMCANot measured38.7%Only Grok 4.20 measured
CL-benchNot measured22.2%Only Grok 4.20 measured
CL-bench LifeNot measured11.9%Only Grok 4.20 measured
LMArena Text1440Not measuredOnly GLM 4.6 measured
LMArena Text: instruction following1410Not measuredOnly GLM 4.6 measured
LMArena Text: creative writing1412Not measuredOnly GLM 4.6 measured
LMArena Text: English1449Not measuredOnly GLM 4.6 measured
LMArena Text: non-English1426Not measuredOnly GLM 4.6 measured
LMArena Text: Chinese1499Not measuredOnly GLM 4.6 measured
LMArena Text: German1447Not measuredOnly GLM 4.6 measured
LMArena Text: Japanese1393Not measuredOnly GLM 4.6 measured
LMArena Text: Korean1400Not measuredOnly GLM 4.6 measured
LMArena Text: French1459Not measuredOnly GLM 4.6 measured
LMArena Text: Spanish1436Not measuredOnly GLM 4.6 measured
LMArena Text: Russian1418Not measuredOnly GLM 4.6 measured
LMArena Text: Polish1447Not measuredOnly GLM 4.6 measured
LMArena Text: multi-turn1427Not measuredOnly GLM 4.6 measured
LMArena Text: longer queries1422Not measuredOnly GLM 4.6 measured
LMArena Text: writing and language1415Not measuredOnly GLM 4.6 measured
LMArena Text: entertainment and media1409Not measuredOnly GLM 4.6 measured
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 4.6 and Grok 4.20
BenchmarkGLM 4.6Grok 4.20Leader
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 4.6$0.0017Fits in one request
Grok 4.20$0.0025Fits in one request

GLM 4.6 costs less

Repository review

50K input + 3K output tokens

GLM 4.6$0.037Fits in one request
Grok 4.20$0.070Fits in one request

GLM 4.6 costs less

Agent loop

200K input + 10K output tokens

GLM 4.6$0.142Exceeds the context window
Grok 4.20$0.275Fits in one request

GLM 4.6 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 4.6 and Grok 4.20
SpecificationZ.ai GLM 4.6SpaceXAI Grok 4.20
MakerZ.aixAI
WeightsOpen weightsNo public weights listed
ReleasedSeptember 30, 2025March 31, 2026
Context window205K2M
Input $/Mtok$0.60$1.25
Output $/Mtok$2.20$2.50
Coding index8.833.3
Independent measurements behind it22
Cost per solved taskNot computable$0.72
SourceOpenRouter listingOpenRouter listing

Questions

Is GLM 4.6 better than Grok 4.20 for coding?

Pick Grok 4.20 for coding: it scores 33.3 on the Vibe Coding Index against 8.8 for GLM 4.6. Pick GLM 4.6 for output tokens 1.1x cheaper ($2.20 vs $2.50 per million) and work measured by Design Arena: agents building mobile apps or Design Arena: agents building Android apps, where it leads.

Which is cheaper, GLM 4.6 or Grok 4.20?

GLM 4.6 costs $0.60 per million input tokens and $2.20 per million output tokens; Grok 4.20 costs $1.25 and $2.50. Cost per solved task is known for Grok 4.20 only.

Which has the bigger context window, GLM 4.6 or Grok 4.20?

Grok 4.20, with 2M tokens against 205K.

How many benchmarks is the GLM 4.6 vs Grok 4.20 comparison based on?

28 benchmarks have a result for both models as of October 21, 2026. GLM 4.6's coding index rests on 2 independent benchmarks and Grok 4.20's on 2. 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-4-6-vs-grok-4-20" title="GLM 4.6 vs Grok 4.20, 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-4-6-vs-grok-4-20">LLMBenchmark.ai, GLM 4.6 vs Grok 4.20</a></p>