Enter opens the highlighted result. Esc closes.

GLM 4.5V vs GLM 4.6V

GLM 4.6V leads GLM 4.5V on the Artificial Analysis Intelligence Index, 11.2 to 7.6, and GLM 4.6V costs 2.0x less per output token ($0.90 vs $1.80 per million). Updated October 21, 2026; 36 benchmarks have a result for both.

Replace GLM 4.5V Replace GLM 4.6V
GLM 4.5VZ.ai
Coding
No data
Overall
No data
Output $/M
$1.80
4
GLM 4.6VZ.ai
Coding
No data
Overall
No data
Output $/M
$0.90
32
vs boards won of 36 shared

GLM 4.5V or GLM 4.6V: which to pick

Neither model has a coding score yet, so the two cannot be ranked on coding.

Pick GLM 4.5V for work measured by Terminal-Bench Hard (Artificial Analysis run) or LMArena Text: coding, where it leads.

Model AGLM 4.5VZ.aiNot benchmarkedNo coding benchmarkNo cost per solved taskModel BGLM 4.6VZ.aiNot benchmarkedNo coding benchmarkNo cost per solved task
Shared results
36
GLM 4.5V only
5
GLM 4.6V only
2
Like-for-like categories
7 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 4.5V not benchmarked, GLM 4.6V not benchmarked.
Widest category
Math, 12.3 points apart on average over 2 shared benchmarks.
Widest single gap
Long context reasoning (AA-LCR): GLM 4.5V 0.0% against GLM 4.6V 48.7%, 48.7 points apart on a 0 to 100 scale.
Output price
GLM 4.5V $1.80 against GLM 4.6V $0.90 per million output tokens, 2.0x apart.
Context window
GLM 4.5V takes 66K tokens, GLM 4.6V 131K.

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 4.5VBarsGLM 4.6VTies
Coding13
Intelligence01
Agentic work01
Reasoning16
Math02
Knowledge05
General214
All shared boards432
CodingLike-for-likeSix of these boards feed the Vibe Coding Index13
Coding benchmarks for GLM 4.5V and GLM 4.6V
BenchmarkGLM 4.5VGLM 4.6VLeader
LiveCodeBench (Artificial Analysis run)60.4%41.1%GLM 4.5V
Terminal-Bench Hard (Artificial Analysis run)6.8%14.4%GLM 4.6V
LMArena Text: coding13471389GLM 4.6V
LMArena Text: software and IT13531391GLM 4.6V
Neither model has a result on the other 35 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index01
Intelligence benchmarks for GLM 4.5V and GLM 4.6V
BenchmarkGLM 4.5VGLM 4.6VLeader
AA Intelligence Index7.611.2GLM 4.6V
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 4.5V and GLM 4.6V
BenchmarkGLM 4.5VGLM 4.6VLeader
τ²-Bench Telecom22.5%31.6%GLM 4.6V
Neither model has a result on the other 40 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index16
Reasoning benchmarks for GLM 4.5V and GLM 4.6V
BenchmarkGLM 4.5VGLM 4.6VLeader
GPQA Diamond (Artificial Analysis run)68.4%71.9%GLM 4.6V
Humanity's Last Exam (Artificial Analysis run)6.3%9.6%GLM 4.6V
LMArena Text: hard prompts13351367GLM 4.6V
LMArena Text: expert13531371GLM 4.6V
LMArena Text: hard prompts in English13571391GLM 4.6V
LMArena Vision: diagrams11511182GLM 4.6V
LMArena Vision: homework11631155GLM 4.5V
Kagi LLM Benchmark59.8%Not measuredOnly GLM 4.5V measured
Neither model has a result on the other 14 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index02
Math benchmarks for GLM 4.5V and GLM 4.6V
BenchmarkGLM 4.5VGLM 4.6VLeader
AA Math Index73.085.3GLM 4.6V
AIME 2025 (Artificial Analysis run)73.0%85.3%GLM 4.6V
LMArena Text: math1354Not measuredOnly GLM 4.5V measured
LMArena Text: mathematical work1336Not measuredOnly GLM 4.5V measured
Neither model has a result on the other 9 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index05
Knowledge benchmarks for GLM 4.5V and GLM 4.6V
BenchmarkGLM 4.5VGLM 4.6VLeader
MMLU Pro (Artificial Analysis run)78.8%79.9%GLM 4.6V
LMArena Text: science13301389GLM 4.6V
LMArena Text: medicine and healthcare13491389GLM 4.6V
LMArena Text: legal and government13381393GLM 4.6V
LMArena Text: business and finance13091359GLM 4.6V
Neither model has a result on the other 13 boards here
Web designNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Web design benchmarks for GLM 4.5V and GLM 4.6V
BenchmarkGLM 4.5VGLM 4.6VLeader
Neither model has a result on 9 boards here
GeneralLike-for-likePublic boards, shown as published, kept out of the index214
General benchmarks for GLM 4.5V and GLM 4.6V
BenchmarkGLM 4.5VGLM 4.6VLeader
IFBench34.2%30.1%GLM 4.5V
Long context reasoning (AA-LCR)0.0%48.7%GLM 4.6V
LMArena Text13331376GLM 4.6V
LMArena Text: instruction following13111350GLM 4.6V
LMArena Text: creative writing12941347GLM 4.6V
LMArena Vision11541164GLM 4.6V
LMArena Text: English13621395GLM 4.6V
LMArena Text: non-English13031358GLM 4.6V
LMArena Text: Chinese13371425GLM 4.6V
LMArena Text: Spanish1336Not measuredOnly GLM 4.5V measured
LMArena Text: Russian12981339GLM 4.6V
LMArena Text: Polish1302Not measuredOnly GLM 4.5V measured
LMArena Text: multi-turn13331359GLM 4.6V
LMArena Text: longer queries13051357GLM 4.6V
LMArena Text: writing and language13021347GLM 4.6V
LMArena Text: entertainment and media13031343GLM 4.6V
LMArena Vision: English11831179GLM 4.5V
LMArena Vision: ChineseNot measured1217Only GLM 4.6V measured
LMArena Vision: reading text11511171GLM 4.6V
LMArena Vision: creative writingNot measured1113Only GLM 4.6V measured
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 4.5V and GLM 4.6V
BenchmarkGLM 4.5VGLM 4.6VLeader
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 4.5V$0.0015Fits in one request
GLM 4.6V$0.0008Fits in one request

GLM 4.6V costs less

Repository review

50K input + 3K output tokens

GLM 4.5V$0.035Fits in one request
GLM 4.6V$0.018Fits in one request

GLM 4.6V costs less

Agent loop

200K input + 10K output tokens

GLM 4.5V$0.138Exceeds the context window
GLM 4.6V$0.069Exceeds the context window

GLM 4.6V 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 4.5V and GLM 4.6V
SpecificationZ.ai GLM 4.5VZ.ai GLM 4.6V
MakerZ.aiZ.ai
WeightsOpen weightsOpen weights
ReleasedAugust 11, 2025December 8, 2025
Context window66K131K
Input $/Mtok$0.60$0.30
Output $/Mtok$1.80$0.90
Coding indexNot benchmarkedNot benchmarked
Independent measurements behind it00
Cost per solved taskNot computableNot computable
SourceOpenRouter listingOpenRouter listing

Questions

Is GLM 4.5V better than GLM 4.6V for coding?

Neither model has a coding score yet, so the two cannot be ranked on coding. Pick GLM 4.5V for work measured by Terminal-Bench Hard (Artificial Analysis run) or LMArena Text: coding, where it leads.

Which is cheaper, GLM 4.5V or GLM 4.6V?

GLM 4.5V costs $0.60 per million input tokens and $1.80 per million output tokens; GLM 4.6V costs $0.30 and $0.90. Cost per solved task is known for neither of them.

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

GLM 4.6V, with 131K tokens against 66K.

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

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

Compare next