Back
Grok

Grok 4.3

xAI, United States170/191 correctLast updated May 20, 2026Website

Grok 4.3 by xAI achieved an overall score of 88.0% on Appwrite Arena, ranking #11 of 12 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 170 of 191 questions correctly across categories including Auth, TablesDB, Functions, Storage, and CLI. During the benchmark it consumed 1,428,904 tokens costing $2.0593, averaging 65.0 tokens per second over 56m 3s. Compare Grok 4.3 with other LLMs on the Appwrite Arena leaderboard.

Total tokens
1,428,904Input: 1,210,404 Output: 218,500
Total duration
56m 3s
Avg speed
65.0 tok/s
Total cost
$ 2.0593
Cost/1MOverallFoundationAuthTablesDBFunctionsStorageSitesMessagingRealtimeCLI
$1.25Input: $1.25/1M tokens
Output: $2.50/1M tokens		88%89.5%87.8%85.7%85%84.3%99%98.2%89.5%77.3%
Deterministic15/17 correct
AI-Judged97% average score