modelgrep

Anthropic: Claude Fable 5 vs MoonshotAI: Kimi K2.7 Code

Anthropic: Claude Fable 5 wins on more metrics (5 of 9), but the right pick depends on what you optimize for — see the breakdown below.

MetricAnthropic: Claude Fable 5MoonshotAI: Kimi K2.7 Code
Intelligence Index59.941.9
Coding Index62.045.6
GPQA Diamond93%90%
Design Arena Elo14151335
Speed (tokens/sec)107
Latency572ms
Input price /M$10.00$0.740
Output price /M$50.00$3.50
Context window1M262K
CapabilitiesReasoningToolsJSONVisionReasoningToolsJSONVision