ClaudevsKimi

Comparing Anthropic and Moonshot AI. Average Podium Score across top models: Claude (84.6) vs Kimi (74.9). Flagship showdown: Claude Mythos Preview (97.4) vs Kimi K3 (83.3).

Claude (Anthropic)

Industry-leading frontier reasoning and coding models from Anthropic.

Avg Score: 84.6
Licensing: Proprietary API

Kimi (Moonshot AI)

High-intelligence reasoning and long-context models from Moonshot AI.

Top Model: Kimi K3
Avg Score: 74.9
Licensing: Proprietary API

Top Models Showdown

RankModelLabScoreCodingReasoningSpeedPrice / 1M
#1Claude Mythos PreviewAnthropic97.480.998.880 t/s$15/M
#2Claude Fable 5Anthropic93.484.181.071 t/s$50/M
#3Kimi K3Moonshot AI83.361.586.537 t/s$15/M
#4Claude Opus 5Anthropic81.470.482.360 t/s$25/M
#5Claude Opus 4.8Anthropic76.058.082.850 t/s$15/M
#6Claude Opus 4 6 ThinkingAnthropic75.050 t/s$15/M
#7Kimi K3 MaxMoonshot AI74.050 t/s$3/M
#8Kimi K2.5 ThinkingMoonshot AI73.050 t/s$3/M
#9Kimi K2.5 InstantMoonshot AI72.050 t/s$3/M
#10Kimi K2 Thinking TurboMoonshot AI72.050 t/s$3/M
0 / 4 Models Selected