SShogo Bench

Claude Fable 5 vs GPT-5.4

Claude Fable 5 leads GPT-5.4 by 12.5 points on the Shogo Agent Index. Claude Fable 5 wins 18 and GPT-5.4 wins 2 of 20 shared benchmarks. GPT-5.4 is cheaper for agent workloads.

Add more models to this comparison →

Pillars

PillarClaude Fable 5GPT-5.4
Agentic coding7457
Long-horizon autonomy6961
Knowledge-work agents5240
Computer use5542
Security & context learning5954

Shared benchmarks

BenchmarkClaude Fable 5GPT-5.4
APEX-Agentsindex63.6%52.4%
DeepSWEindex69.9%51.8%
Vending-Bench 2index$5,680$6,144
GSO-Benchindex78.4%31.4%
PostTrainBenchindex37.3%17.8%
ARC-AGI98.5%93.7%
ARC-AGI-289.2%74.0%
Chess Puzzles47.0%48.0%
CritPt28.6%23.4%
DTBench98.4%94.4%
FrontierMath Tier 490.2%49.0%
FrontierMath Tiers 1-387.0%78.6%
GPQA Diamond93.9%93.3%
LMCA61.2%52.0%
MirrorCode70.0%23.3%
OTIS Mock AIME100.0%97.8%
ProofBench95.0%56.0%
SciCode61.0%56.6%
SimpleQA Verified70.7%45.1%
WebDev Arena16281463

Specs and price

Claude Fable 5GPT-5.4
Agent Index64.1 (#12)51.6 (#33)
Agent $/M$4.62$1.27
Input / output $/M$10.00 / $50.00$2.50 / $15.00
Context1M1.1M
ReleasedJun 9, 2026Mar 5, 2026
WeightsClosedClosed
API hosts44

Don’t pick one. Route.

Shogo can run Claude Fable 5 and GPT-5.4 side by side and send each step of an agent to the one that’s cheaper for it.

Try both in Shogo