SShogo Bench

GPT-5.6 Sol vs Muse Spark 1.3

GPT-5.6 Sol leads Muse Spark 1.3 by 7.1 points on the Shogo Agent Index, but their ranges overlap, so treat it as close. GPT-5.6 Sol wins 10 and Muse Spark 1.3 wins 7 of 17 shared benchmarks. Muse Spark 1.3 is cheaper for agent workloads.

Add more models to this comparison →

Pillars

PillarGPT-5.6 SolMuse Spark 1.3
Agentic coding7160
Long-horizon autonomy7264
Knowledge-work agents4645
Computer use5343
Security & context learning5853

Shared benchmarks

BenchmarkGPT-5.6 SolMuse Spark 1.3
APEX-Agentsindex51.4%58.6%
FrontierSWEindex32.2%25.8%
CursorBenchindex41.7%41.6%
Chess Puzzles68.0%37.0%
CritPt32.3%26.0%
DTBench96.0%96.5%
FrontierMath Tier 482.9%46.3%
FrontierMath Tiers 1-389.1%74.4%
GeoBench73.0%86.0%
LMCA59.2%53.9%
MirrorCode20.0%54.4%
OTIS Mock AIME100.0%99.2%
ProofBench83.0%58.0%
SciCode57.8%59.7%
SimpleBench71.7%81.8%
WebDev Arena16171652
WeirdML v315.0%6.7%

Specs and price

GPT-5.6 SolMuse Spark 1.3
Agent Index61.8 (#14)54.7 (#28)
Agent $/M$1.85$0.030
Input / output $/M$4.00 / $20.00$0.10 / $0.20
Context1.1M1M
ReleasedAug 7, 2026Sep 2, 2026
WeightsClosedClosed
API hosts52

Don’t pick one. Route.

Shogo can run GPT-5.6 Sol 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