SShogo Bench

Kimi K3 vs Muse Spark 1.3

Kimi K3 leads Muse Spark 1.3 by 0.2 points on the Shogo Agent Index, but their ranges overlap, so treat it as close. Kimi K3 wins 4 and Muse Spark 1.3 wins 9 of 15 shared benchmarks. Muse Spark 1.3 is cheaper for agent workloads.

Add more models to this comparison →

Pillars

PillarKimi K3Muse Spark 1.3
Agentic coding6360
Long-horizon autonomy6564
Knowledge-work agents4245
Computer use4343
Security & context learning5353

Shared benchmarks

BenchmarkKimi K3Muse Spark 1.3
APEX-Agentsindex47.8%58.6%
FrontierSWEindex25.9%25.8%
Chess Puzzles37.0%37.0%
CritPt23.4%26.0%
DTBench91.2%96.5%
FrontierMath Tier 439.0%46.3%
FrontierMath Tiers 1-372.2%74.4%
LMCA52.7%53.9%
MirrorCode54.4%54.4%
OTIS Mock AIME97.2%99.2%
ProofBench87.0%58.0%
SciCode59.5%59.7%
SimpleBench60.7%81.8%
WebDev Arena16741652
WeirdML v37.5%6.7%

Specs and price

Kimi K3Muse Spark 1.3
Agent Index54.9 (#26)54.7 (#28)
Agent $/M$1.39$0.030
Input / output $/M$3.00 / $15.00$0.10 / $0.20
Context1M1M
ReleasedJul 16, 2026Sep 2, 2026
WeightsOpenClosed
API hosts292

Don’t pick one. Route.

Shogo can run Kimi K3 and Muse Spark 1.3 side by side and send each step of an agent to the one that’s cheaper for it.

Try both in Shogo