SShogo Bench

GPT-5.4 vs Muse Spark 1.3

Muse Spark 1.3 leads GPT-5.4 by 3.1 points on the Shogo Agent Index, but their ranges overlap, so treat it as close. GPT-5.4 wins 3 and Muse Spark 1.3 wins 9 of 12 shared benchmarks. Muse Spark 1.3 is cheaper for agent workloads.

Add more models to this comparison →

Pillars

PillarGPT-5.4Muse Spark 1.3
Agentic coding5760
Long-horizon autonomy6164
Knowledge-work agents4045
Computer use4243
Security & context learning5453

Shared benchmarks

BenchmarkGPT-5.4Muse Spark 1.3
APEX-Agentsindex52.4%58.6%
Chess Puzzles48.0%37.0%
CritPt23.4%26.0%
DTBench94.4%96.5%
FrontierMath Tier 449.0%46.3%
FrontierMath Tiers 1-378.6%74.4%
LMCA52.0%53.9%
MirrorCode23.3%54.4%
OTIS Mock AIME97.8%99.2%
ProofBench56.0%58.0%
SciCode56.6%59.7%
WebDev Arena14631652

Specs and price

GPT-5.4Muse Spark 1.3
Agent Index51.6 (#33)54.7 (#28)
Agent $/M$1.27$0.030
Input / output $/M$2.50 / $15.00$0.10 / $0.20
Context1.1M1M
ReleasedMar 5, 2026Sep 2, 2026
WeightsClosedClosed
API hosts42

Don’t pick one. Route.

Shogo can run GPT-5.4 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