SShogo Bench

GLM-5.3-Flash vs Kimi K3

Kimi K3 leads GLM-5.3-Flash by 6.6 points on the Shogo Agent Index, but their ranges overlap, so treat it as close. GLM-5.3-Flash wins 3 and Kimi K3 wins 13 of 16 shared benchmarks. GLM-5.3-Flash is cheaper for agent workloads.

Add more models to this comparison →

Pillars

PillarGLM-5.3-FlashKimi K3
Agentic coding5463
Long-horizon autonomy5765
Knowledge-work agents4042
Computer use3443
Security & context learning4753

Shared benchmarks

BenchmarkGLM-5.3-FlashKimi K3
APEX-Agentsindex52.8%47.8%
DeepSWEindex63.4%68.5%
FrontierSWEindex18.1%25.9%
PostTrainBenchindex36.8%31.5%
ARC-AGI91.0%94.5%
ARC-AGI-265.8%60.4%
Chess Puzzles21.0%37.0%
CritPt15.4%23.4%
FrontierCode31.8%44.2%
FrontierMath Tier 417.1%39.0%
FrontierMath Tiers 1-355.8%72.2%
GPQA Diamond90.2%93.1%
OTIS Mock AIME93.9%97.2%
ProofBench21.0%87.0%
SciCode51.6%59.5%
WebDev Arena16071674

Specs and price

GLM-5.3-FlashKimi K3
Agent Index48.3 (#38)54.9 (#26)
Agent $/M$0.16$1.39
Input / output $/M$0.30 / $1.05$3.00 / $15.00
Context1M1M
ReleasedSep 18, 2026Jul 16, 2026
WeightsOpenOpen
API hosts4229

Don’t pick one. Route.

Shogo can run GLM-5.3-Flash 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