SShogo Bench

Claude Haiku 5.5 vs GPT-6.1 Terra

GPT-6.1 Terra leads Claude Haiku 5.5 by 11.5 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 Terra wins 11 of 12 shared benchmarks. Claude Haiku 5.5 is cheaper for agent workloads.

Add more models to this comparison →

Pillars

PillarClaude Haiku 5.5GPT-6.1 Terra
Agentic coding6779
Long-horizon autonomy6274
Knowledge-work agents4656
Computer use4763
Security & context learning5562

Shared benchmarks

BenchmarkClaude Haiku 5.5GPT-6.1 Terra
APEX-Agentsindex56.1%68.3%
Vending-Bench 2index$3,602$9,107
CritPt22.6%28.0%
FrontierMath Tier 446.3%92.7%
FrontierMath Tiers 1-375.1%90.9%
GPQA Diamond89.6%94.3%
MirrorCode78.9%76.7%
OTIS Mock AIME98.9%100.0%
SciCode55.0%55.1%
SimpleBench59.2%72.2%
SimpleQA Verified23.8%57.3%
WeirdML v311.7%28.1%

Specs and price

Claude Haiku 5.5GPT-6.1 Terra
Agent Index57.5 (#18)69 (#8)
Agent $/M$0.050$0.42
Input / output $/M$0.10 / $0.50$1.00 / $5.00
Context1M1.1M
ReleasedOct 7, 2026Oct 26, 2026
WeightsClosedClosed
API hosts54

Don’t pick one. Route.

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

Try both in Shogo