LLMBenchmark.ai

Enter opens the highlighted result. Esc closes.

GLM 5 vs Muse Spark 1.1

Updated October 21, 2026. 58 benchmarks have a result for both. Coding is ranked on the Vibe Coding Index; every other board is shown as published.

Replace GLM 5 Replace Muse Spark 1.1
GLM 5Z.ai
Coding
59.7
Overall
27.9
Output $/M
$3.20
2
Muse Spark 1.1Meta
Coding
48.9
Overall
33.7
Output $/M
$4.25
56
vs boards won of 58 shared

Decision reading

Pick GLM 5 for coding: it scores 59.7 on the Vibe Coding Index against 48.9 for Muse Spark 1.1.

Pick Muse Spark 1.1 when a solved task costs $0.43 against $0.63, or when the context window is larger, 1.05M against 205K tokens.

Model AGLM 5Z.ai59.7/1002 independent benchmarks$0.63 per solved taskModel BMuse Spark 1.1Meta48.9/1001 independent benchmark$0.43 per solved task
Shared results
58
GLM 5 only
17
Muse Spark 1.1 only
38
Like-for-like categories
8 of 9

Shared results can support a head-to-head reading. Results only one model has describe coverage, not superiority.

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 59.7 from 2 benchmarks, Muse Spark 1.1 48.9 from 1 benchmark.
Widest category
Coding, 15.5 points apart on average over 8 shared benchmarks.
Widest single gap
Vibe Code Bench (Vals AI): GLM 5 23.4% against Muse Spark 1.1 72.2%, 48.8 points apart on a 0 to 100 scale.
Output price
GLM 5 $3.20 against Muse Spark 1.1 $4.25 per million output tokens, 1.3x apart.
Context window
GLM 5 takes 205K tokens, Muse Spark 1.1 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 5BarsMuse Spark 1.1Ties
Coding08
Intelligence02
Agentic work15
Reasoning06
Math02
Knowledge08
Web design08
General117
All shared boards256
CodingLike-for-likeFour of these boards feed the Vibe Coding Index08
Coding benchmarks for GLM 5 and Muse Spark 1.1
BenchmarkGLM 5Muse Spark 1.1Leader
SWE-bench Verified72.8%Not measuredOnly GLM 5 measured
SWE-bench Multilingual69.7%Not measuredOnly GLM 5 measured
LMArena WebDev14351542Muse Spark 1.1
AA Coding IndexNot measured71.3Only Muse Spark 1.1 measured
LiveCodeBench (Vals AI)81.9%85.9%Muse Spark 1.1
SWE-bench Verified (Vals AI)71.4%82.0%Muse Spark 1.1
Vibe Code Bench (Vals AI)23.4%72.2%Muse Spark 1.1
Code migration (Vals AI)Not measured31.1%Only Muse Spark 1.1 measured
ProgramBench (Vals AI)Not measured0.0%Only Muse Spark 1.1 measured
Terminal-Bench 2.1 (Vals AI)Not measured69.3%Only Muse Spark 1.1 measured
SkillsBench (Vals AI)Not measured59.2%Only Muse Spark 1.1 measured
SciCodeNot measured58.8%Only Muse Spark 1.1 measured
Terminal-Bench Hard (Artificial Analysis run)43.2%Not measuredOnly GLM 5 measured
Terminal-Bench52.4%Not measuredOnly GLM 5 measured
WeirdML48.2%Not measuredOnly GLM 5 measured
ALE-Bench765.6Not measuredOnly GLM 5 measured
LMArena Text: coding14611498Muse Spark 1.1
LMArena Text: software and IT14601498Muse Spark 1.1
LMArena WebDev: React14241532Muse Spark 1.1
LMArena WebDev: HTML14501540Muse Spark 1.1
LMArena WebDev: from an imageNot measured1540Only Muse Spark 1.1 measured
Neither model has a result on the other 17 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index02
Intelligence benchmarks for GLM 5 and Muse Spark 1.1
BenchmarkGLM 5Muse Spark 1.1Leader
AA Intelligence Index27.933.7Muse Spark 1.1
Epoch Capabilities Index145.8154.1Muse Spark 1.1
Agentic workLike-for-likePublic boards, shown as published, kept out of the index15
Agentic work benchmarks for GLM 5 and Muse Spark 1.1
BenchmarkGLM 5Muse Spark 1.1Leader
AA Agentic IndexNot measured25.8Only Muse Spark 1.1 measured
Design Arena: agents building web appsNot measured1207Only Muse Spark 1.1 measured
Design Arena: agents building full-stack apps11231199Muse Spark 1.1
Design Arena: agents building mobile apps11361177Muse Spark 1.1
Design Arena: agents building Android apps11521184Muse Spark 1.1
Design Arena: agents building Godot games11441103GLM 5
Design Arena: agents building gamesNot measured1183Only Muse Spark 1.1 measured
Design Arena: HTML slides11481194Muse Spark 1.1
Design Arena: PowerPoint slidesNot measured1156Only Muse Spark 1.1 measured
Design Arena: spreadsheetsNot measured974Only Muse Spark 1.1 measured
Finance Agent (Vals AI)Not measured57.2%Only Muse Spark 1.1 measured
Legal Research Bench (Vals AI)Not measured38.0%Only Muse Spark 1.1 measured
Excel Modeling Benchmark (Vals AI)Not measured56.4%Only Muse Spark 1.1 measured
Time Horizon Index (Vals AI)Not measured1.7%Only Muse Spark 1.1 measured
τ²-Bench Telecom98.2%Not measuredOnly GLM 5 measured
APEX-AgentsNot measured31.8%Only Muse Spark 1.1 measured
Vending-Bench 2$4,432$6,520Muse Spark 1.1
Neither model has a result on the other 22 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index06
Reasoning benchmarks for GLM 5 and Muse Spark 1.1
BenchmarkGLM 5Muse Spark 1.1Leader
GPQA Diamond83.3%91.2%Muse Spark 1.1
MMMU Pro (Vals AI)Not measured86.6%Only Muse Spark 1.1 measured
GPQA Diamond (Artificial Analysis run)82.0%89.8%Muse Spark 1.1
Humanity's Last Exam (Artificial Analysis run)29.3%46.2%Muse Spark 1.1
GPQA Diamond (Epoch AI run)87.8%Not measuredOnly GLM 5 measured
ARC-AGI44.7%Not measuredOnly GLM 5 measured
ARC-AGI-24.9%Not measuredOnly GLM 5 measured
SimpleBench53.2%Not measuredOnly GLM 5 measured
Chess puzzlesNot measured34.0%Only Muse Spark 1.1 measured
LMArena Text: hard prompts14521485Muse Spark 1.1
LMArena Text: expert14541478Muse Spark 1.1
LMArena Text: hard prompts in English14601480Muse Spark 1.1
LMArena Vision: diagramsNot measured1305Only Muse Spark 1.1 measured
LMArena Vision: homeworkNot measured1287Only Muse Spark 1.1 measured
Neither model has a result on the other 7 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index02
Math benchmarks for GLM 5 and Muse Spark 1.1
BenchmarkGLM 5Muse Spark 1.1Leader
AIME91.7%Not measuredOnly GLM 5 measured
OTIS Mock AIME80.0%Not measuredOnly GLM 5 measured
ProofBenchNot measured39.0%Only Muse Spark 1.1 measured
LMArena Text: math14401482Muse Spark 1.1
LMArena Text: mathematical work14511487Muse Spark 1.1
Neither model has a result on the other 8 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index08
Knowledge benchmarks for GLM 5 and Muse Spark 1.1
BenchmarkGLM 5Muse Spark 1.1Leader
MMLU Pro86.0%88.7%Muse Spark 1.1
LegalBench (Vals AI)84.1%85.0%Muse Spark 1.1
CaseLaw (Vals AI)52.5%Not measuredOnly GLM 5 measured
CorpFin (Vals AI)62.9%71.3%Muse Spark 1.1
TaxEval (Vals AI)70.0%79.7%Muse Spark 1.1
MortgageTax (Vals AI)Not measured66.1%Only Muse Spark 1.1 measured
MedQA (Vals AI)94.3%Not measuredOnly GLM 5 measured
SimpleQA VerifiedNot measured57.8%Only Muse Spark 1.1 measured
LMArena Text: science14601484Muse Spark 1.1
LMArena Text: medicine and healthcare14551479Muse Spark 1.1
LMArena Text: legal and government14531481Muse Spark 1.1
LMArena Text: business and finance14301470Muse Spark 1.1
Neither model has a result on the other 5 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index08
Web design benchmarks for GLM 5 and Muse Spark 1.1
BenchmarkGLM 5Muse Spark 1.1Leader
Design Arena: websites12471270Muse Spark 1.1
Design Arena: UI components12301290Muse Spark 1.1
Design Arena: game dev12341269Muse Spark 1.1
Design Arena: data viz12281268Muse Spark 1.1
Design Arena: 3D12291260Muse Spark 1.1
Design Arena: all code categories12461276Muse Spark 1.1
Design Arena: SVG11681234Muse Spark 1.1
Design Arena: ASCII art11461270Muse Spark 1.1
Neither model has a result on the other 1 board here
GeneralLike-for-likePublic boards, shown as published, kept out of the index117
General benchmarks for GLM 5 and Muse Spark 1.1
BenchmarkGLM 5Muse Spark 1.1Leader
Public Benefits Bench (Vals AI)Not measured61.3%Only Muse Spark 1.1 measured
SAGE (Vals AI)Not measured45.6%Only Muse Spark 1.1 measured
MedScribe (Vals AI)Not measured88.9%Only Muse Spark 1.1 measured
Vals Multimodal IndexNot measured66.7%Only Muse Spark 1.1 measured
IFBench72.3%Not measuredOnly GLM 5 measured
Long context reasoning (AA-LCR)75.7%77.7%Muse Spark 1.1
DTBenchNot measured94.4%Only Muse Spark 1.1 measured
LMCANot measured49.9%Only Muse Spark 1.1 measured
CL-bench18.7%Not measuredOnly GLM 5 measured
Surface Evolver BenchNot measured52.5%Only Muse Spark 1.1 measured
LMArena Text14461478Muse Spark 1.1
LMArena Text: instruction following14281457Muse Spark 1.1
LMArena Text: creative writing14381435GLM 5
LMArena VisionNot measured1293Only Muse Spark 1.1 measured
LMArena Text: English14581478Muse Spark 1.1
LMArena Text: non-English14301471Muse Spark 1.1
LMArena Text: Chinese15111515Muse Spark 1.1
LMArena Text: German14471469Muse Spark 1.1
LMArena Text: Japanese14161450Muse Spark 1.1
LMArena Text: Korean14231459Muse Spark 1.1
LMArena Text: French14561492Muse Spark 1.1
LMArena Text: Spanish14541464Muse Spark 1.1
LMArena Text: Russian14361482Muse Spark 1.1
LMArena Text: Polish14341499Muse Spark 1.1
LMArena Text: multi-turn14561484Muse Spark 1.1
LMArena Text: longer queries14461462Muse Spark 1.1
LMArena Text: writing and language14371448Muse Spark 1.1
LMArena Text: entertainment and media14261438Muse Spark 1.1
LMArena Vision: EnglishNot measured1292Only Muse Spark 1.1 measured
LMArena Vision: ChineseNot measured1348Only Muse Spark 1.1 measured
LMArena Vision: reading textNot measured1301Only Muse Spark 1.1 measured
LMArena Vision: humorNot measured1288Only Muse Spark 1.1 measured
LMArena Vision: creative writingNot measured1253Only Muse Spark 1.1 measured
LMArena DocumentNot measured1465Only Muse Spark 1.1 measured
Neither model has a result on the other 12 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for GLM 5 and Muse Spark 1.1
BenchmarkGLM 5Muse Spark 1.1Leader
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$0.0026Fits in one request
Muse Spark 1.1$0.0034Fits in one request

GLM 5 costs less

Repository review

50K input + 3K output tokens

GLM 5$0.060Fits in one request
Muse Spark 1.1$0.075Fits in one request

GLM 5 costs less

Agent loop

200K input + 10K output tokens

GLM 5$0.232Exceeds the context window
Muse Spark 1.1$0.292Fits in one request

GLM 5 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 and Muse Spark 1.1
SpecificationZ.ai GLM 5Meta Muse Spark 1.1
MakerZ.aiMeta
WeightsOpen weightsNo public weights listed
ReleasedFebruary 11, 2026July 16, 2026
Context window205K1.05M
Input $/Mtok$1.00$1.25
Output $/Mtok$3.20$4.25
Coding index59.748.9
Independent measurements behind it21
Cost per solved task$0.63$0.43
SourceOpenRouter listingOpenRouter listing

Questions

Is GLM 5 better than Muse Spark 1.1 for coding?

Pick GLM 5 for coding: it scores 59.7 on the Vibe Coding Index against 48.9 for Muse Spark 1.1. Pick Muse Spark 1.1 when a solved task costs $0.43 against $0.63, or when the context window is larger, 1.05M against 205K tokens.

Which is cheaper, GLM 5 or Muse Spark 1.1?

GLM 5 costs $1.00 per million input tokens and $3.20 per million output tokens; Muse Spark 1.1 costs $1.25 and $4.25. Per solved SWE-bench Verified task, GLM 5 costs $0.63 and Muse Spark 1.1 $0.43.

Which has the bigger context window, GLM 5 or Muse Spark 1.1?

Muse Spark 1.1, with 1.05M tokens against 205K.

How many benchmarks is the GLM 5 vs Muse Spark 1.1 comparison based on?

58 benchmarks have a result for both models as of October 21, 2026. GLM 5's coding index rests on 2 independent benchmarks and Muse Spark 1.1's on 1. Where only one model has a result, the page says so instead of naming a winner.

Compare next