SShogo Bench

Claude Fable 5.2 vs Claude Opus 5.5

Claude Fable 5.2 leads Claude Opus 5.5 by 5.5 points on the Shogo Agent Index, but their ranges overlap, so treat it as close. Claude Fable 5.2 wins 18 and Claude Opus 5.5 wins 1 of 21 shared benchmarks. Claude Opus 5.5 is cheaper for agent workloads.

Add more models to this comparison →

Pillars

PillarClaude Fable 5.2Claude Opus 5.5
Agentic coding8681
Long-horizon autonomy7876
Knowledge-work agents6758
Computer use7766
Security & context learning6562

Shared benchmarks

BenchmarkClaude Fable 5.2Claude Opus 5.5
APEX-Agentsindex80.1%73.5%
Vending-Bench 2index$9,628$9,235
FrontierSWEindex69.9%62.3%
Remote Labor Indexindex41.3%21.3%
CursorBenchindex60.0%57.8%
PostTrainBenchindex44.6%43.8%
ARC-AGI98.5%98.5%
ARC-AGI-293.8%93.3%
Chess Puzzles79.0%67.0%
CritPt35.7%31.7%
DTBench99.2%98.9%
FrontierMath Tier 4100.0%95.0%
FrontierMath Tiers 1-395.1%91.2%
GPQA Diamond96.0%94.1%
LMCA70.6%68.2%
OTIS Mock AIME99.7%100.0%
ProofBench100.0%100.0%
SciCode67.7%66.9%
SimpleBench94.9%88.4%
SimpleQA Verified94.0%72.2%
WeirdML v343.1%31.2%

Specs and price

Claude Fable 5.2Claude Opus 5.5
Agent Index76.4 (#1)70.9 (#3)
Agent $/M$3.21$1.69
Input / output $/M$8.00 / $40.00$4.00 / $20.00
Context1M1M
ReleasedNov 9, 2026Sep 22, 2026
WeightsClosedClosed
API hosts46

Don’t pick one. Route.

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

Try both in Shogo