SShogo Bench

Claude Opus 4.7 vs Kimi K3

Kimi K3 leads Claude Opus 4.7 by 0.3 points on the Shogo Agent Index, but their ranges overlap, so treat it as close. Claude Opus 4.7 wins 8 and Kimi K3 wins 12 of 20 shared benchmarks. Kimi K3 is cheaper for agent workloads.

Add more models to this comparison →

Pillars

PillarClaude Opus 4.7Kimi K3
Agentic coding6063
Long-horizon autonomy6965
Knowledge-work agents4142
Computer use4443
Security & context learning5253

Shared benchmarks

BenchmarkClaude Opus 4.7Kimi K3
APEX-Agentsindex49.2%47.8%
Vending-Bench 2index$10,937$5,165
PostTrainBenchindex24.3%31.5%
ARC-AGI93.5%94.5%
ARC-AGI-275.8%60.4%
Chess Puzzles41.0%37.0%
CritPt12.0%23.4%
DTBench94.7%91.2%
FrontierCode38.5%44.2%
FrontierMath Tier 431.7%39.0%
FrontierMath Tiers 1-370.2%72.2%
GPQA Diamond90.2%93.1%
LMCA52.2%52.7%
MirrorCode43.3%54.4%
OTIS Mock AIME97.8%97.2%
ProofBench54.0%87.0%
SciCode54.5%59.5%
SimpleBench61.7%60.7%
SimpleQA Verified51.7%50.6%
WebDev Arena15671674

Specs and price

Claude Opus 4.7Kimi K3
Agent Index54.6 (#29)54.9 (#26)
Agent $/M$2.31$1.39
Input / output $/M$5.00 / $25.00$3.00 / $15.00
Context1M1M
ReleasedApr 16, 2026Jul 16, 2026
WeightsClosedOpen
API hosts529

Don’t pick one. Route.

Shogo can run Claude Opus 4.7 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