SShogo Bench

GPT-6 Sol vs Muse Wonder

GPT-6 Sol leads Muse Wonder by 2.7 points on the Shogo Agent Index, but their ranges overlap, so treat it as close. GPT-6 Sol wins 9 and Muse Wonder wins 3 of 12 shared benchmarks. GPT-6 Sol is cheaper for agent workloads.

Add more models to this comparison →

Pillars

PillarGPT-6 SolMuse Wonder
Agentic coding7268
Long-horizon autonomy7670
Knowledge-work agents4850
Computer use5450
Security & context learning5957

Shared benchmarks

BenchmarkGPT-6 SolMuse Wonder
APEX-Agentsindex54.3%66.1%
Vending-Bench 2index$14,428$9,432
CritPt30.9%28.3%
DTBench97.3%95.2%
FrontierMath Tier 490.0%70.7%
FrontierMath Tiers 1-389.8%82.1%
LMCA59.1%56.8%
MirrorCode48.9%45.6%
OTIS Mock AIME100.0%99.4%
ProofBench83.0%84.0%
SciCode57.6%60.3%
WeirdML v319.7%15.1%

Specs and price

GPT-6 SolMuse Wonder
Agent Index63.6 (#13)60.9 (#15)
Agent $/M$0.92$1.69
Input / output $/M$2.00 / $10.00$4.00 / $20.00
Context1.1M1M
ReleasedSep 22, 2026Oct 12, 2026
WeightsClosedClosed
API hosts52

Don’t pick one. Route.

Shogo can run GPT-6 Sol and Muse Wonder side by side and send each step of an agent to the one that’s cheaper for it.

Try both in Shogo