SShogo Bench

GPT-5.5 vs Kimi K3

GPT-5.5 leads Kimi K3 by 1.8 points on the Shogo Agent Index, but their ranges overlap, so treat it as close. GPT-5.5 wins 14 and Kimi K3 wins 7 of 21 shared benchmarks. Kimi K3 is cheaper for agent workloads.

Add more models to this comparison →

Pillars

PillarGPT-5.5Kimi K3
Agentic coding6363
Long-horizon autonomy6765
Knowledge-work agents4642
Computer use4343
Security & context learning5553

Shared benchmarks

BenchmarkGPT-5.5Kimi K3
APEX-Agentsindex55.1%47.8%
DeepSWEindex67.0%68.5%
Vending-Bench 2index$7,524$5,165
PostTrainBenchindex23.9%31.5%
ARC-AGI95.0%94.5%
ARC-AGI-285.0%60.4%
Chess Puzzles63.0%37.0%
CritPt27.1%23.4%
DTBench96.0%91.2%
FrontierCode43.0%44.2%
FrontierMath Tier 472.5%39.0%
FrontierMath Tiers 1-385.3%72.2%
GPQA Diamond94.0%93.1%
LMCA54.3%52.7%
MirrorCode23.3%54.4%
OTIS Mock AIME100.0%97.2%
ProofBench50.0%87.0%
SciCode56.1%59.5%
SimpleBench69.0%60.7%
SimpleQA Verified63.0%50.6%
WebDev Arena15101674

Specs and price

GPT-5.5Kimi K3
Agent Index56.7 (#19)54.9 (#26)
Agent $/M$2.55$1.39
Input / output $/M$5.00 / $30.00$3.00 / $15.00
Context1.1M1M
ReleasedApr 23, 2026Jul 16, 2026
WeightsClosedOpen
API hosts429

Don’t pick one. Route.

Shogo can run GPT-5.5 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