SShogo Bench

Claude Haiku 5.5 vs Muse Wonder

Muse Wonder leads Claude Haiku 5.5 by 3.4 points on the Shogo Agent Index, but their ranges overlap, so treat it as close. Claude Haiku 5.5 wins 2 and Muse Wonder wins 11 of 13 shared benchmarks. Claude Haiku 5.5 is cheaper for agent workloads.

Add more models to this comparison →

Pillars

PillarClaude Haiku 5.5Muse Wonder
Agentic coding6768
Long-horizon autonomy6270
Knowledge-work agents4650
Computer use4750
Security & context learning5557

Shared benchmarks

BenchmarkClaude Haiku 5.5Muse Wonder
APEX-Agentsindex56.1%66.1%
Vending-Bench 2index$3,602$9,432
CursorBenchindex48.4%50.4%
CritPt22.6%28.3%
DTBench95.0%95.2%
FrontierMath Tier 446.3%70.7%
FrontierMath Tiers 1-375.1%82.1%
LMCA51.1%56.8%
MirrorCode78.9%45.6%
OTIS Mock AIME98.9%99.4%
ProofBench89.0%84.0%
SciCode55.0%60.3%
WeirdML v311.7%15.1%

Specs and price

Claude Haiku 5.5Muse Wonder
Agent Index57.5 (#18)60.9 (#15)
Agent $/M$0.050$1.69
Input / output $/M$0.10 / $0.50$4.00 / $20.00
Context1M1M
ReleasedOct 7, 2026Oct 12, 2026
WeightsClosedClosed
API hosts52

Don’t pick one. Route.

Shogo can run Claude Haiku 5.5 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