SShogo Bench

Claude Haiku 5.5 vs GPT-6.1 Sol

GPT-6.1 Sol leads Claude Haiku 5.5 by 12.5 points on the Shogo Agent Index, but their ranges overlap, so treat it as close. Claude Haiku 5.5 wins 1 and GPT-6.1 Sol wins 18 of 19 shared benchmarks. Claude Haiku 5.5 is cheaper for agent workloads.

Add more models to this comparison →

Pillars

PillarClaude Haiku 5.5GPT-6.1 Sol
Agentic coding6780
Long-horizon autonomy6278
Knowledge-work agents4654
Computer use4766
Security & context learning5562

Shared benchmarks

BenchmarkClaude Haiku 5.5GPT-6.1 Sol
APEX-Agentsindex56.1%60.0%
Vending-Bench 2index$3,602$12,861
ARC-AGI89.0%98.5%
ARC-AGI-261.3%94.2%
Chess Puzzles55.0%68.0%
CritPt22.6%31.7%
DTBench95.0%96.5%
FrontierCode46.4%50.2%
FrontierMath Tier 446.3%100.0%
FrontierMath Tiers 1-375.1%93.7%
GPQA Diamond89.6%95.4%
LMCA51.1%62.8%
MirrorCode78.9%61.1%
OTIS Mock AIME98.9%100.0%
ProofBench89.0%99.0%
SciCode55.0%55.8%
SimpleBench59.2%82.9%
SimpleQA Verified23.8%73.9%
WeirdML v311.7%38.3%

Specs and price

Claude Haiku 5.5GPT-6.1 Sol
Agent Index57.5 (#18)70 (#5)
Agent $/M$0.050$0.84
Input / output $/M$0.10 / $0.50$2.00 / $10.00
Context1M1.1M
ReleasedOct 7, 2026Sep 29, 2026
WeightsClosedClosed
API hosts56

Don’t pick one. Route.

Shogo can run Claude Haiku 5.5 and GPT-6.1 Sol side by side and send each step of an agent to the one that’s cheaper for it.

Try both in Shogo