SShogo Bench

Claude Fable 5.1 vs GPT-5.6 Luna

Claude Fable 5.1 leads GPT-5.6 Luna by 16.8 points on the Shogo Agent Index. Claude Fable 5.1 wins 20 and GPT-5.6 Luna wins 0 of 20 shared benchmarks. GPT-5.6 Luna is cheaper for agent workloads.

Add more models to this comparison →

Pillars

PillarClaude Fable 5.1GPT-5.6 Luna
Agentic coding8059
Long-horizon autonomy7161
Knowledge-work agents5539
Computer use6240
Security & context learning6151

Shared benchmarks

BenchmarkClaude Fable 5.1GPT-5.6 Luna
APEX-Agentsindex68.6%43.0%
Vending-Bench 2index$5,422$4,095
CursorBenchindex51.8%35.9%
ARC-AGI97.5%88.0%
ARC-AGI-290.0%59.5%
Chess Puzzles48.0%43.0%
CritPt31.1%20.6%
DTBench97.6%89.1%
FrontierCode50.9%39.8%
FrontierMath Tier 487.8%61.0%
FrontierMath Tiers 1-390.2%82.1%
LMCA65.5%48.5%
MirrorCode93.3%21.1%
OTIS Mock AIME100.0%98.3%
ProofBench100.0%60.0%
SciCode63.1%53.6%
SimpleBench86.6%46.8%
SimpleQA Verified70.8%41.0%
WebDev Arena17581519
WeirdML v326.0%5.2%

Specs and price

Claude Fable 5.1GPT-5.6 Luna
Agent Index68.2 (#10)51.4 (#34)
Agent $/M$4.01$0.10
Input / output $/M$10.00 / $50.00$0.20 / $1.20
Context1M1.1M
ReleasedSep 1, 2026Jul 9, 2026
WeightsClosedClosed
API hosts35

Don’t pick one. Route.

Shogo can run Claude Fable 5.1 and GPT-5.6 Luna side by side and send each step of an agent to the one that’s cheaper for it.

Try both in Shogo