SShogo Bench

Claude Haiku 5.5 vs Kimi K3

Claude Haiku 5.5 leads Kimi K3 by 2.6 points on the Shogo Agent Index, but their ranges overlap, so treat it as close. Claude Haiku 5.5 wins 12 and Kimi K3 wins 8 of 20 shared benchmarks. Claude Haiku 5.5 is cheaper for agent workloads.

Add more models to this comparison →

Pillars

PillarClaude Haiku 5.5Kimi K3
Agentic coding6763
Long-horizon autonomy6265
Knowledge-work agents4642
Computer use4743
Security & context learning5553

Shared benchmarks

BenchmarkClaude Haiku 5.5Kimi K3
APEX-Agentsindex56.1%47.8%
Vending-Bench 2index$3,602$5,165
FrontierSWEindex43.8%25.9%
ARC-AGI89.0%94.5%
ARC-AGI-261.3%60.4%
Chess Puzzles55.0%37.0%
CritPt22.6%23.4%
DTBench95.0%91.2%
FrontierCode46.4%44.2%
FrontierMath Tier 446.3%39.0%
FrontierMath Tiers 1-375.1%72.2%
GPQA Diamond89.6%93.1%
LMCA51.1%52.7%
MirrorCode78.9%54.4%
OTIS Mock AIME98.9%97.2%
ProofBench89.0%87.0%
SciCode55.0%59.5%
SimpleBench59.2%60.7%
SimpleQA Verified23.8%50.6%
WeirdML v311.7%7.5%

Specs and price

Claude Haiku 5.5Kimi K3
Agent Index57.5 (#18)54.9 (#26)
Agent $/M$0.050$1.39
Input / output $/M$0.10 / $0.50$3.00 / $15.00
Context1M1M
ReleasedOct 7, 2026Jul 16, 2026
WeightsClosedOpen
API hosts529

Don’t pick one. Route.

Shogo can run Claude Haiku 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