SShogo Bench

GLM-5.3 vs Muse Spark 1.3

GLM-5.3 leads Muse Spark 1.3 by 0.5 points on the Shogo Agent Index, but their ranges overlap, so treat it as close. GLM-5.3 wins 3 and Muse Spark 1.3 wins 10 of 14 shared benchmarks. Muse Spark 1.3 is cheaper for agent workloads.

Add more models to this comparison →

Pillars

PillarGLM-5.3Muse Spark 1.3
Agentic coding6260
Long-horizon autonomy6864
Knowledge-work agents4345
Computer use4043
Security & context learning5153

Shared benchmarks

BenchmarkGLM-5.3Muse Spark 1.3
APEX-Agentsindex56.6%58.6%
FrontierSWEindex30.2%25.8%
CursorBenchindex42.6%41.6%
Chess Puzzles37.0%37.0%
CritPt19.1%26.0%
DTBench87.7%96.5%
FrontierMath Tier 429.3%46.3%
FrontierMath Tiers 1-368.8%74.4%
LMCA55.5%53.9%
OTIS Mock AIME91.1%99.2%
ProofBench49.0%58.0%
SciCode59.0%59.7%
SimpleBench66.2%81.8%
WebDev Arena16141652

Specs and price

GLM-5.3Muse Spark 1.3
Agent Index55.2 (#25)54.7 (#28)
Agent $/M$0.61$0.030
Input / output $/M$1.20 / $4.18$0.10 / $0.20
Context1M1M
ReleasedAug 14, 2026Sep 2, 2026
WeightsOpenClosed
API hosts492

Don’t pick one. Route.

Shogo can run GLM-5.3 and Muse Spark 1.3 side by side and send each step of an agent to the one that’s cheaper for it.

Try both in Shogo