llm.ing

Kimi K3

Moonshot AIopen weightsOpenRouter ↗

Multimodal Kimi model with 1M context and toggleable max-effort thinking for long-horizon agent work

Benchmarks

BenchmarkCategoryScoreVariantProvenanceSourceObserved
AA Agentic Indexagentic50.1mirroropenrouterJul 23, 2026
AA Coding Indexcoding76.2mirroropenrouterJul 23, 2026
AA Intelligence Indexintelligence57.1mirroropenrouterJul 23, 2026
Design Arena — 3dpreference1465mirroropenrouterJul 24, 2026
Design Arena — codecategoriespreference1431mirroropenrouterJul 24, 2026
Design Arena — datavizpreference1373mirroropenrouterJul 24, 2026
Design Arena — godotgamedevpreference1199mirroropenrouterJul 23, 2026
Design Arena — python-pptxslidespreference1280mirroropenrouterJul 24, 2026
Design Arena — uicomponentpreference1431mirroropenrouterJul 24, 2026
Design Arena — webappspreference1313mirroropenrouterJul 24, 2026
Design Arena — websitepreference1371mirroropenrouterJul 24, 2026
LMArena Textpreference1482crowdlmarenaJul 21, 2026
LMArena WebDevpreference1678crowdlmarenaJul 21, 2026

Includes index data from Artificial Analysis.

Providers

Live operational stats per hosting endpoint, via OpenRouter.

ProviderQuantContext$/1M in · outUptime 24hLatencyThroughput
Moonshot AIint41.0M$3 · $1599.99%