SShogo Bench

Claude Fable 5 vs Kimi K3

Claude Fable 5 leads Kimi K3 by 9.2 points on the Shogo Agent Index, but their ranges overlap, so treat it as close. Claude Fable 5 wins 22 and Kimi K3 wins 1 of 23 shared benchmarks. Kimi K3 is cheaper for agent workloads.

Add more models to this comparison →

Pillars

PillarClaude Fable 5Kimi K3
Agentic coding7463
Long-horizon autonomy6965
Knowledge-work agents5242
Computer use5543
Security & context learning5953

Shared benchmarks

BenchmarkClaude Fable 5Kimi K3
APEX-Agentsindex63.6%47.8%
DeepSWEindex69.9%68.5%
Vending-Bench 2index$5,680$5,165
FrontierSWEindex47.0%25.9%
PostTrainBenchindex37.3%31.5%
ARC-AGI98.5%94.5%
ARC-AGI-289.2%60.4%
Chess Puzzles47.0%37.0%
CritPt28.6%23.4%
DTBench98.4%91.2%
FrontierCode53.5%44.2%
FrontierMath Tier 490.2%39.0%
FrontierMath Tiers 1-387.0%72.2%
GPQA Diamond93.9%93.1%
LMCA61.2%52.7%
MirrorCode70.0%54.4%
OTIS Mock AIME100.0%97.2%
ProofBench95.0%87.0%
SciCode61.0%59.5%
SimpleBench81.9%60.7%
SimpleQA Verified70.7%50.6%
WebDev Arena16281674
WeirdML v315.6%7.5%

Specs and price

Claude Fable 5Kimi K3
Agent Index64.1 (#12)54.9 (#26)
Agent $/M$4.62$1.39
Input / output $/M$10.00 / $50.00$3.00 / $15.00
Context1M1M
ReleasedJun 9, 2026Jul 16, 2026
WeightsClosedOpen
API hosts429

Don’t pick one. Route.

Shogo can run Claude Fable 5 and Kimi K3 side by side and send each step of an agent to the one that’s cheaper for it.

Try both in Shogo