SShogo Bench

Claude Haiku 5.5 vs GPT-5.6 Sol

GPT-5.6 Sol leads Claude Haiku 5.5 by 4.3 points on the Shogo Agent Index, but their ranges overlap, so treat it as close. Claude Haiku 5.5 wins 5 and GPT-5.6 Sol wins 16 of 21 shared benchmarks. Claude Haiku 5.5 is cheaper for agent workloads.

Add more models to this comparison →

Pillars

PillarClaude Haiku 5.5GPT-5.6 Sol
Agentic coding6771
Long-horizon autonomy6272
Knowledge-work agents4646
Computer use4753
Security & context learning5558

Shared benchmarks

BenchmarkClaude Haiku 5.5GPT-5.6 Sol
APEX-Agentsindex56.1%51.4%
Vending-Bench 2index$3,602$9,619
FrontierSWEindex43.8%32.2%
CursorBenchindex48.4%41.7%
ARC-AGI89.0%97.5%
ARC-AGI-261.3%92.5%
Chess Puzzles55.0%68.0%
CritPt22.6%32.3%
DTBench95.0%96.0%
FrontierCode46.4%47.5%
FrontierMath Tier 446.3%82.9%
FrontierMath Tiers 1-375.1%89.1%
GPQA Diamond89.6%93.5%
LMCA51.1%59.2%
MirrorCode78.9%20.0%
OTIS Mock AIME98.9%100.0%
ProofBench89.0%83.0%
SciCode55.0%57.8%
SimpleBench59.2%71.7%
SimpleQA Verified23.8%69.7%
WeirdML v311.7%15.0%

Specs and price

Claude Haiku 5.5GPT-5.6 Sol
Agent Index57.5 (#18)61.8 (#14)
Agent $/M$0.050$1.85
Input / output $/M$0.10 / $0.50$4.00 / $20.00
Context1M1.1M
ReleasedOct 7, 2026Aug 7, 2026
WeightsClosedClosed
API hosts55

Don’t pick one. Route.

Shogo can run Claude Haiku 5.5 and GPT-5.6 Sol side by side and send each step of an agent to the one that’s cheaper for it.

Try both in Shogo