SShogo Bench

Claude Fable 5.2 vs GPT-5.6 Luna

Claude Fable 5.2 leads GPT-5.6 Luna by 25.0 points on the Shogo Agent Index. Claude Fable 5.2 wins 18 and GPT-5.6 Luna wins 0 of 18 shared benchmarks. GPT-5.6 Luna is cheaper for agent workloads.

Add more models to this comparison →

Pillars

PillarClaude Fable 5.2GPT-5.6 Luna
Agentic coding8659
Long-horizon autonomy7861
Knowledge-work agents6739
Computer use7740
Security & context learning6551

Shared benchmarks

BenchmarkClaude Fable 5.2GPT-5.6 Luna
APEX-Agentsindex80.1%43.0%
Vending-Bench 2index$9,628$4,095
CursorBenchindex60.0%35.9%
ARC-AGI98.5%88.0%
ARC-AGI-293.8%59.5%
Chess Puzzles79.0%43.0%
CritPt35.7%20.6%
DTBench99.2%89.1%
FrontierMath Tier 4100.0%61.0%
FrontierMath Tiers 1-395.1%82.1%
GPQA Diamond96.0%91.6%
LMCA70.6%48.5%
OTIS Mock AIME99.7%98.3%
ProofBench100.0%60.0%
SciCode67.7%53.6%
SimpleBench94.9%46.8%
SimpleQA Verified94.0%41.0%
WeirdML v343.1%5.2%

Specs and price

Claude Fable 5.2GPT-5.6 Luna
Agent Index76.4 (#1)51.4 (#34)
Agent $/M$3.21$0.10
Input / output $/M$8.00 / $40.00$0.20 / $1.20
Context1M1.1M
ReleasedNov 9, 2026Jul 9, 2026
WeightsClosedClosed
API hosts45

Don’t pick one. Route.

Shogo can run Claude Fable 5.2 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