Back
Grok

Grok 4.6

xAI, United States188/191 correctLast updated Aug 24, 2026Website

Grok 4.6 by xAI achieved an overall score of 97.8% on Appwrite Arena, ranking #2 of 24 benchmarked models. The benchmark tests how well AI models understand Appwrite - the open-source backend platform for authentication, TablesDB, storage, functions, and more. This model answered 188 of 191 questions correctly across categories including Auth, TablesDB, Functions, Storage, and CLI. During the benchmark it consumed 1,530,418 tokens costing $3.7710, averaging 49.0 tokens per second over 60m 21s. Compare Grok 4.6 with other LLMs on the Appwrite Arena leaderboard.

Total tokens
1,530,418Input: 1,352,868 Output: 177,550
Total duration
60m 21s
Avg speed
49.0 tok/s
Total cost
$ 3.7710
Cost/1MOverallFoundationAuthTablesDBFunctionsStorageSitesMessagingRealtimeCLI
$2.00Input: $2.00/1M tokens
Output: $6.00/1M tokens		97.8%100%99.3%100%98.8%94.5%100%95%95%97.7%
Deterministic17/17 correct
AI-Judged100% average score