SShogo Bench

Claude Fable 5.2 vs GPT-5.5

Claude Fable 5.2 leads GPT-5.5 by 19.7 points on the Shogo Agent Index. Claude Fable 5.2 wins 17 and GPT-5.5 wins 1 of 18 shared benchmarks. GPT-5.5 is cheaper for agent workloads.

Add more models to this comparison →

Pillars

PillarClaude Fable 5.2GPT-5.5
Agentic coding8663
Long-horizon autonomy7867
Knowledge-work agents6746
Computer use7743
Security & context learning6555

Shared benchmarks

BenchmarkClaude Fable 5.2GPT-5.5
APEX-Agentsindex80.1%55.1%
Vending-Bench 2index$9,628$7,524
Remote Labor Indexindex41.3%6.3%
PostTrainBenchindex44.6%23.9%
ARC-AGI98.5%95.0%
ARC-AGI-293.8%85.0%
Chess Puzzles79.0%63.0%
CritPt35.7%27.1%
DTBench99.2%96.0%
FrontierMath Tier 4100.0%72.5%
FrontierMath Tiers 1-395.1%85.3%
GPQA Diamond96.0%94.0%
LMCA70.6%54.3%
OTIS Mock AIME99.7%100.0%
ProofBench100.0%50.0%
SciCode67.7%56.1%
SimpleBench94.9%69.0%
SimpleQA Verified94.0%63.0%

Specs and price

Claude Fable 5.2GPT-5.5
Agent Index76.4 (#1)56.7 (#19)
Agent $/M$3.21$2.55
Input / output $/M$8.00 / $40.00$5.00 / $30.00
Context1M1.1M
ReleasedNov 9, 2026Apr 23, 2026
WeightsClosedClosed
API hosts44

Don’t pick one. Route.

Shogo can run Claude Fable 5.2 and GPT-5.5 side by side and send each step of an agent to the one that’s cheaper for it.

Try both in Shogo