SShogo Bench

Claude Fable 5.1 vs Kimi K3

Claude Fable 5.1 leads Kimi K3 by 13.3 points on the Shogo Agent Index. Claude Fable 5.1 wins 21 and Kimi K3 wins 0 of 21 shared benchmarks. Kimi K3 is cheaper for agent workloads.

Add more models to this comparison →

Pillars

PillarClaude Fable 5.1Kimi K3
Agentic coding8063
Long-horizon autonomy7165
Knowledge-work agents5542
Computer use6243
Security & context learning6153

Shared benchmarks

BenchmarkClaude Fable 5.1Kimi K3
APEX-Agentsindex68.6%47.8%
Vending-Bench 2index$5,422$5,165
FrontierSWEindex56.3%25.9%
PostTrainBenchindex44.6%31.5%
ARC-AGI97.5%94.5%
ARC-AGI-290.0%60.4%
Chess Puzzles48.0%37.0%
CritPt31.1%23.4%
DTBench97.6%91.2%
FrontierCode50.9%44.2%
FrontierMath Tier 487.8%39.0%
FrontierMath Tiers 1-390.2%72.2%
LMCA65.5%52.7%
MirrorCode93.3%54.4%
OTIS Mock AIME100.0%97.2%
ProofBench100.0%87.0%
SciCode63.1%59.5%
SimpleBench86.6%60.7%
SimpleQA Verified70.8%50.6%
WebDev Arena17581674
WeirdML v326.0%7.5%

Specs and price

Claude Fable 5.1Kimi K3
Agent Index68.2 (#10)54.9 (#26)
Agent $/M$4.01$1.39
Input / output $/M$10.00 / $50.00$3.00 / $15.00
Context1M1M
ReleasedSep 1, 2026Jul 16, 2026
WeightsClosedOpen
API hosts329

Don’t pick one. Route.

Shogo can run Claude Fable 5.1 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