Back
Kimi

Kimi K2.5

MoonshotAI, China177/191 correctLast updated Mar 13, 2026Website

Kimi K2.5 by MoonshotAI achieved an overall score of 91.7% on Appwrite Arena, ranking #5 of 10 benchmarked models. The benchmark tests how well AI models understand Appwrite - the open-source backend platform for authentication, databases, storage, functions, and more. This model answered 177 of 191 questions correctly across categories including Auth, Databases, Functions, Storage, and CLI. During the benchmark it consumed 1,563,978 tokens costing $0.8642, averaging 35.3 tokens per second over 43m 18s. Compare Kimi K2.5 with other LLMs on the Appwrite Arena leaderboard.

Total tokens
1,563,978Input: 1,472,333 Output: 91,645
Total duration
43m 18s
Avg speed
35.3 tok/s
Total cost
$ 0.8642
Cost/1MOverallFoundationAuthDatabasesFunctionsStorageSitesMessagingRealtimeCLI
$0.45Input: $0.45/1M tokens
Output: $2.20/1M tokens		91.7%99.3%97.5%84.1%93%89%99.5%99.5%78%87.9%
Deterministic17/17 correct
AI-Judged95% average score