Enter opens the highlighted result. Esc closes.

GLM 4.5 vs GLM 4.6

GLM 4.6 leads GLM 4.5 on the Artificial Analysis Intelligence Index, 18.5 to 12.8, at about the same output price ($2.20 vs $2.20 per million). Updated October 21, 2026; 58 benchmarks have a result for both.

Replace GLM 4.5 Replace GLM 4.6
GLM 4.5Z.ai
Coding
No data
Overall
No data
Output $/M
$2.20
15
GLM 4.6Z.ai
Coding
8.8
Overall
33.4
Output $/M
$2.20
43
vs boards won of 58 shared

GLM 4.5 or GLM 4.6: which to pick

Only GLM 4.6 has a coding score, 8.8 on the Vibe Coding Index; GLM 4.5 has none, so the two cannot be ranked on coding.

Pick GLM 4.5 for work measured by LiveCodeBench (Artificial Analysis run) or ALE-Bench, where it leads.

Model AGLM 4.5Z.aiNot benchmarkedNo coding benchmarkNo cost per solved taskModel BGLM 4.6Z.ai8.8/1002 independent benchmarksNo cost per solved task
Shared results
58
GLM 4.5 only
3
GLM 4.6 only
11
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 4.5 not benchmarked, GLM 4.6 8.8 from 2 benchmarks.
Widest category
Agentic work, 33.9 points apart on average over 1 shared benchmark.
Widest single gap
τ²-Bench Telecom: GLM 4.5 43.0% against GLM 4.6 76.9%, 33.9 points apart on a 0 to 100 scale.
Output price
Both cost $2.20 per million output tokens.
Context window
GLM 4.5 takes 131K tokens, GLM 4.6 205K.

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 4.6Ties
Coding25
Intelligence01
Agentic work01
Reasoning34
Math15
Knowledge55
Web design16
General316
All shared boards1543
CodingLike-for-likeSix of these boards feed the Vibe Coding Index25
Coding benchmarks for GLM 4.5 and GLM 4.6
BenchmarkGLM 4.5GLM 4.6Leader
SWE-bench Verified64.2%68.2%GLM 4.6
LMArena WebDevNot measured1341Only GLM 4.6 measured
AA Coding IndexNot measured45.8Only GLM 4.6 measured
LiveCodeBench (Vals AI)67.5%81.0%GLM 4.6
Vibe Code Bench (Vals AI)Not measured3.1%Only GLM 4.6 measured
LiveCodeBench (Artificial Analysis run)73.8%69.5%GLM 4.5
Terminal-Bench Hard (Artificial Analysis run)22.0%28.8%GLM 4.6
Terminal-BenchNot measured24.5%Only GLM 4.6 measured
WeirdML40.6%Not measuredOnly GLM 4.5 measured
ALE-Bench344.8340.8GLM 4.5
LMArena Text: coding14341449GLM 4.6
LMArena Text: software and IT14401455GLM 4.6
LMArena WebDev: HTMLNot measured1350Only GLM 4.6 measured
Neither model has a result on the other 26 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index01
Intelligence benchmarks for GLM 4.5 and GLM 4.6
BenchmarkGLM 4.5GLM 4.6Leader
AA Intelligence Index12.818.5GLM 4.6
Neither model has a result on the other 1 board here
Agentic workLike-for-likePublic boards, shown as published, kept out of the index01
Agentic work benchmarks for GLM 4.5 and GLM 4.6
BenchmarkGLM 4.5GLM 4.6Leader
Design Arena: agents building full-stack appsNot measured1029Only GLM 4.6 measured
Design Arena: agents building mobile appsNot measured1096Only GLM 4.6 measured
Design Arena: agents building Android appsNot measured1084Only GLM 4.6 measured
Design Arena: agents building Godot gamesNot measured1180Only GLM 4.6 measured
τ²-Bench Telecom43.0%76.9%GLM 4.6
Berkeley Function Calling Leaderboard (BFCL)Not measured72.4%Only GLM 4.6 measured
BrowseComp-PlusNot measured48.2%Only GLM 4.6 measured
Neither model has a result on the other 34 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index34
Reasoning benchmarks for GLM 4.5 and GLM 4.6
BenchmarkGLM 4.5GLM 4.6Leader
GPQA Diamond72.2%74.5%GLM 4.6
GPQA Diamond (Artificial Analysis run)78.2%78.0%GLM 4.5
Humanity's Last Exam (Artificial Analysis run)13.0%14.5%GLM 4.6
LMArena Text: hard prompts14291440GLM 4.6
LMArena Text: expert14331432GLM 4.5
LMArena Text: hard prompts in English14291445GLM 4.6
Kagi LLM Benchmark57.9%45.7%GLM 4.5
Neither model has a result on the other 15 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index15
Math benchmarks for GLM 4.5 and GLM 4.6
BenchmarkGLM 4.5GLM 4.6Leader
AIME86.7%92.7%GLM 4.6
MATH 500 (Vals AI)94.0%Not measuredOnly GLM 4.5 measured
MGSM (Vals AI)90.8%89.8%GLM 4.5
AA Math Index73.786.0GLM 4.6
AIME 2025 (Artificial Analysis run)73.7%86.0%GLM 4.6
LMArena Text: math14271432GLM 4.6
LMArena Text: mathematical work14301447GLM 4.6
Neither model has a result on the other 6 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index55
Knowledge benchmarks for GLM 4.5 and GLM 4.6
BenchmarkGLM 4.5GLM 4.6Leader
MMLU Pro81.2%82.2%GLM 4.6
LegalBench (Vals AI)75.6%79.6%GLM 4.6
CorpFin (Vals AI)61.0%56.8%GLM 4.5
TaxEval (Vals AI)72.4%66.2%GLM 4.5
MedQA (Vals AI)90.0%92.2%GLM 4.6
MMLU Pro (Artificial Analysis run)83.5%82.9%GLM 4.5
LMArena Text: science14441456GLM 4.6
LMArena Text: medicine and healthcare14551455GLM 4.5
LMArena Text: legal and government14381448GLM 4.6
LMArena Text: business and finance14201416GLM 4.5
Neither model has a result on the other 8 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index16
Web design benchmarks for GLM 4.5 and GLM 4.6
BenchmarkGLM 4.5GLM 4.6Leader
Design Arena: websites11701174GLM 4.6
Design Arena: UI components11441148GLM 4.6
Design Arena: game dev11451151GLM 4.6
Design Arena: data viz11591161GLM 4.6
Design Arena: 3D11701129GLM 4.5
Design Arena: all code categories11651166GLM 4.6
Design Arena: SVG10961109GLM 4.6
Neither model has a result on the other 2 boards here
GeneralLike-for-likePublic boards, shown as published, kept out of the index316
General benchmarks for GLM 4.5 and GLM 4.6
BenchmarkGLM 4.5GLM 4.6Leader
IFBench44.1%43.4%GLM 4.5
Long context reasoning (AA-LCR)52.7%54.0%GLM 4.6
Creative writing (Lech Mazur)73.4%Not measuredOnly GLM 4.5 measured
LMArena Text14301440GLM 4.6
LMArena Text: instruction following14041410GLM 4.6
LMArena Text: creative writing13951412GLM 4.6
LMArena Text: English14371449GLM 4.6
LMArena Text: non-English14171426GLM 4.6
LMArena Text: Chinese14651499GLM 4.6
LMArena Text: German14071447GLM 4.6
LMArena Text: Japanese14151393GLM 4.5
LMArena Text: Korean13801400GLM 4.6
LMArena Text: French14181459GLM 4.6
LMArena Text: Spanish14541436GLM 4.5
LMArena Text: Russian14131418GLM 4.6
LMArena Text: Polish14221447GLM 4.6
LMArena Text: multi-turn14151427GLM 4.6
LMArena Text: longer queries14121422GLM 4.6
LMArena Text: writing and language14031415GLM 4.6
LMArena Text: entertainment and media14041409GLM 4.6
Neither model has a result on the other 26 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 4.6
BenchmarkGLM 4.5GLM 4.6Leader
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 4.6$0.0017Fits in one request

Same cost

Repository review

50K input + 3K output tokens

GLM 4.5$0.037Fits in one request
GLM 4.6$0.037Fits in one request

Same cost

Agent loop

200K input + 10K output tokens

GLM 4.5$0.142Exceeds the context window
GLM 4.6$0.142Exceeds the context window

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 4.5 and GLM 4.6
SpecificationZ.ai GLM 4.5Z.ai GLM 4.6
MakerZ.aiZ.ai
WeightsOpen weightsOpen weights
ReleasedJuly 25, 2025September 30, 2025
Context window131K205K
Input $/Mtok$0.60$0.60
Output $/Mtok$2.20$2.20
Coding indexNot benchmarked8.8
Independent measurements behind it02
Cost per solved taskNot computableNot computable
SourceOpenRouter listingOpenRouter listing

Questions

Is GLM 4.5 better than GLM 4.6 for coding?

Only GLM 4.6 has a coding score, 8.8 on the Vibe Coding Index; GLM 4.5 has none, so the two cannot be ranked on coding. Pick GLM 4.5 for work measured by LiveCodeBench (Artificial Analysis run) or ALE-Bench, where it leads.

Which is cheaper, GLM 4.5 or GLM 4.6?

GLM 4.5 costs $0.60 per million input tokens and $2.20 per million output tokens; GLM 4.6 costs $0.60 and $2.20. Cost per solved task is known for neither of them.

Which has the bigger context window, GLM 4.5 or GLM 4.6?

GLM 4.6, with 205K tokens against 131K.

How many benchmarks is the GLM 4.5 vs GLM 4.6 comparison based on?

58 benchmarks have a result for both models as of October 21, 2026. GLM 4.5's coding index rests on 0 independent benchmarks and GLM 4.6's on 2. Where only one model has a result, the page says so instead of naming a winner.

Compare next